jim walsh - kirsch foundation · lessons from success: the npt and the future of non-proliferation...

no 41

Learning from Past Success:The NPT and the Future of Non-proliferation

JIM WALSH

THE WEAPONS OF MASS DESTRUCTION COMMISSION

www.wmdcommission.org

This paper has been commissioned by the Weapons of Mass Destruction Commission. Its purpose is to function as food-for-thought for the work of the Commission. The Commission is not responsible for views expressed in this paper.

Weapons of Mass Destruction Commission (WMDC)The WMDC is an independent international commission initiated by the Swedish Government on a proposal from the United Nations. Its task is to present proposals aimed at the greatest possible reduction of the dangers of weapons of mass destruction, including both short-term and long-term approaches and both non-proliferation and disarmament aspects. The Commission will issue its report in early 2006.

The commissioners serve in their personal capacity. The Commission is supported by a Secretariat based in Stockholm, Sweden.

Members of the Weapons of Mass Destruction CommissionHans Blix, Chairman (Sweden) Dewi Fortuna Anwar (Indonesia)Alexei G Arbatov (Russian Federation)Marcos de Azambuja (Brazil)Alyson Bailes (United Kingdom)Jayantha Dhanapala (Sri Lanka)Gareth Evans (Australia)Patricia Lewis (Ireland)Masashi Nishihara (Japan)William J. Perry (United States of America)Vasantha Raghavan (India)Cheikh Sylla (Senegal)Prince El Hassan bin Talal (Jordan)Pan, Zhenqiang (China)

Secretary-General of the CommissionHenrik Salander (Sweden)

Weapons of Mass Destruction CommissionPostal address: SE-103 33 Stockholm, SwedenVisitors’ address: Regeringsgatan 30–32E-mail: [email protected]: www.wmdcommission.org

Learning from Past Success:

The NPT and the Future of Non-proliferation

Jim Walsh, Harvard University

Paper prepared for the Weapons of Mass Destruction Commission

Stockholm, Sweden

October, 2005

Lessons from Success: The NPT and the Future of Non-proliferationJim Walsh,

p. 2

Table of Contents

I. Introduction: Three Questions

II. Recognizing and Understanding Success: Why Success Matters

III. NPT: Success or Failure?

A. The Conventional Wisdom

B. Evidence and Arguments for Success and Failure

C. Measuring Success and Failure

D. Unappreciated Success

IV. What Accounts for the Success of the NPT Regime?

A. The Puzzle

B. Explaining the Puzzle: Potential Explanations

C. Conclusion: A Missing Element

V. The Primacy of Politics: NPT’s Influence on Internal Decision Making

A. Traditional Non-proliferation Analysis: Abstraction and Automaticity

B. A Political Model of Nuclear Decision Making

C. The Impact of the NPT on Internal Nuclear Decision Making

VI. Lessons (Un)Learned: Past Success and Today’s Non-proliferation Challenges

A. Today’s Non-proliferation Challenges

B. Lessons from the NPT for Non-proliferation Policy in General

C. Lessons Applied to Today’s Non-proliferation Challenges

D. Limits to the Lessons

VII. Conclusion: The Three Questions


p. 3

I. Introduction: Three Questions

This paper addresses three questions.1 First, has the Nuclear Non-proliferation Treaty (NPT)

been a success or a failure? Second, what accounts for this success or failure? Third, what do

the lessons from the first thirty-five years of the treaty suggest, if anything, about how to address

the problem of proliferation and in particular, the post-Cold War and post-9/11 challenges that

confront the international non-proliferation regime?

The analysis presented here finds that the NPT has been surprisingly successful. Indeed, it is

arguably the most successful arms control treaty in human history. Unfortunately—and

strangely—this success has been either ignored or discounted as irrelevant. It will be argued that

failure to fully acknowledge and understand the success of the NPT is a potentially dangerous

policy error. To understand the treaty’s success, the paper looks beyond security and technical

factors and instead considers political factors and the way in which the NPT influenced

countries’ internal decision dynamics. It concludes by looking at the lessons learned and their

implications for non-proliferation policy.

Before turning to the question of whether the treaty was a success or a failure, it makes sense to

first consider the meaning and importance of “success.”

1 This paper is based on two talks given at meetings sponsored by the Weapons of MassDestruction Commission, one in Helsinki, Finland in March of 2005 (also supported by theFinnish Ministry of Foreign Affairs) and another at the NPT Review Conference in May of thatsame year. The author is indebted to the Commission for its support and to the many colleaguesat these and related meetings who offered their thoughts. In particular, Manne Wangborg andAnthony Wier were generous with their comments and ideas.


p. 4

II. Recognizing and Understanding Success: Why Success Matters

There are at least two reasons why attention to policy success is important. The first is that a

focus on success yields certain policy benefits. Policy makers who confront a problem need to

make choices from an array of possible remedies and in an environment of limited time, limited

resources, and limited information (uncertainty). Policy successes provide decision makers with

instruments that are proven to be effective and more certain than untested alternatives. In

addition, building on already established success is usually politically easier and more efficient

than pursing new initiatives. This is not to suggest that governments should avoid new policy

remedies. Rather, the point is that decision makers should first consider what has already

worked and whether it makes sense to build on that success.

A second reason for understanding why success is important is that failure to do so entails policy

risks. A failure to appreciate the extent or frequency of success can lead officials to exaggerate

dangers and opt for extreme measures where costs may be high and where effectiveness is

doubtful. Even extreme measures may look attractive, however, if dangers appear dire. With

respect to proliferation, for example, advocates for missile defence, counter proliferation, and

preventive war have argued that the failure of the non-proliferation regime and the inevitability

of proliferation make their policies the preferable choice.

Another kind of risk comes from a failure to understand the causes of success. Ignoring the

causes of success leaves decision makers vulnerable to taking actions that undermine the reasons

or conditions for previous success, thus making a bad situation even worse. The United States,

for example, has been tempted to pursue a new generation of “bunker busting” or “mini” nuclear

weapons in order to deter so-called rogue states.2 However, such policies intended to contribute

2 Michael Klare, Rogue States and Nuclear Outlaws, (New York: Hill and Wang, 1995); AshokKapur, "Rogue States and the International Nuclear Order," International Journal, 51:3 (Summer1996), pp. 420-439; Mary Caprioli, and Peter F. Trumbore, “Rhetoric Versus Reality: RogueStates in Interstate Conflict,” Journal of Conflict Resolution, 49:5 (October 2005), pp. 770-791.


p. 5

to the general collapse of the non-proliferation regime, may instead result in contribution to

proliferation.

A third risk involves systems whose effectiveness depends, at least in part, on the confidence of

its members. The NPT is just such a system. The treaty is stronger when member states believe

that the treaty is working and that nations can be counted on to abide by the rules. It is weaker

when governments believe that the treaty is not working or failing. A perception of failure

encourages states to consider alternatives such as hedging and gives pro-bomb advocates an

opening to make their case. This is not to suggest that discussions of the NPT should ignore

failures and engage in boosterism. Focusing only on failure is misleading, however, and runs the

risk of generating a spiralling crisis of confidence. In such a spiral, a government begins to pull

back from its commitments because of doubts about the treaty, which in turn is taken as evidence

that the treaty is weak and encourages other governments to follow suit.

Ignoring success thus carries many risks. Of course, exaggerating success can be dangerous as

well. Rather than ignoring or hyping success, what is really required is an objective assessment

of the regime. In the next section, the question of success and failure is examined in depth. It

begins by examining a point of view that is widely held if not predominant. It then proceeds

with a review of the historical record and concludes by considering the standards one might use

for measuring success.

III. NPT: Success or Failure?

A. The Conventional Wisdom

For most observers and commentators, there is little debate regarding the success or failure of the

NPT: there has been a long running, broad consensus that the NPT a) has failed, b) is on the

verge of failure, or c) will inevitably fail with a resulting cascade of proliferation. 3 Mary Thornton, “Is Non-proliferation in Jeopardy? Recommendations for U.S. NuclearWeapons Policy,” JPIA: Journal of Public and International Affairs, 13 (Spring 2002), pp. 181-


p. 6

Sometimes the statements are explicit, as when some commentators have baldly asserted that the

NPT has “failed” or is “collapsing.”4 As one observer recently noted, “It has become

fashionable, yet again, to predict the collapse of the NPT…. The last time such a debate

happened was in the aftermath of the Indian and Pakistani nuclear tests in the summer of 1998.

Now, North Korea's apparent breakout of the NPT Treaty, and Iran's challenge to the NPT

regime are leading to yet another debate about the future of the regime.”5 Even the office of the

UN Secretary-General has suggested that “we are approaching a point at which the erosion of the

non-proliferation regime could become irreversible and result in a cascade of proliferation.”6

At other times, the logic of failure is implicit. Typically, it takes the form of predictions that

nuclear weapons will proliferate despite the presence of the treaty.7 Prior to the NPT, there were

numerous and infamous predictions of widespread proliferation8, and when the treaty was

199.4 Those using the word “fail”include William C. Martel, “The End of Non-proliferation?,”Strategic Review. 28:4 (Fall 2000), pp. 16-21 and Jason D. Ellis, “The Best Defense:Counterproliferation and U.S. National Security,” Washington Quarterly, 26:2 (Spring 2003), pp.119-120, but it is the more moderate William Pfaff who suggested that “the NPT is collapsing.”William Pfaff, “The Nuclear Realities,” The Advertiser, September 10, 2005, p. W3. See alsoMichael MccGwire, “The Rise and Fall of the NPT: An Opportunity for Britain,” InternationalAffairs, 81:1 (January 2005), pp. 115-140. Not surprisingly, analysts in India, Pakistan, andIsrael, who do not want to be punished or constrained by their absence in the NPT, have offeredsimilar views. See for example, Narayanan Komerath, “The Nuclear Truth Triumphs,” SecurityResearch Review, 2005, Bharat-Rakshak,http://www.bharatrakshak.com/SRR/Volume14/komerath.html; C. Raja Mohan, “On NuclearDiplomacy, a Window Has Opened,” Indian Express, October 19, 2004,http://www.observerindia.com/analysis/A293.htm.5 Rajesh Rajagopalan, “Will the NPT Regime Survive?,” ORF Strategic Trends, 1:2 (October 6,2003), Observer Research Foundation.6 A More Secure World: Our Shared Responsibility, Report of the Secretary-General’s High-level Panel on Threats, Challenges and Change, (New York: United Nations, 2004), pp. 40.7 Obviously, a prediction that nuclear weapons will spread is a prediction the NPT regime willfail.8 The most widely cited erroneous prediction was John F. Kennedy’s but his view was broadlyrepresentative. See Public Papers of the President of the United States: John F. Kennedy, 1963(Washington, D.C.: Government Printing Office, 1964), p. 280 or New York Times, March 23,1963, p. 1. Draft Memorandum for the President from the Secretary of State, "The Diffusion of


p. 7

presented for consideration, doubts were plentiful. One anonymous official spoke for many

when he lamented that…

The Treaty appears in some ways to be a negation of history. Allpeople with the knowledge and resources they needed haveprogressed through evolutions and revolutions in industry,transport and weapons; from the manual to the machine, fromsailing ships to steamers, from the oxcart to the aeroplane, andfrom the club to gun and bomb.9

Despite these worries, the NPT was adopted. What is striking, however, is that predictions of

widespread proliferation continued with consistency. Such predictions – as well as benedictions

for a dying regime – were most frequent in the aftermath of perceived proliferation crises.10

After the Indian test in 1974, the Iraq revelations of the early 1990s11, the North Korean stand-off

Nuclear Weapons with and without a Test Ban Agreement," July 27, 1962, Table 1, John F.Kennedy Presidential Library. See also Sir John Cockcroft, “The Perils of NuclearProliferation,” in Unless Peace Comes, in Unless Peace Comes, Nigel Calder, ed., (New York:Viking Press, 1968), p. 37. Proliferation, wrote Gallois, was “as irreversible as . . . thegeneralization of firearms.” Pierre Gallois, The Balance of Terror, (Boston: Houghton Mifflin,1961), p. 229.9 X. “Australian Doubts on the Treaty,” Quadrant, 12:53 (May-June 1968), p. 31.10 Predictions of NPT collapse and surging proliferation have not limited to proliferation crises,however. With the end of the Cold War, several scholars of international relations predicted thatthe end of the superpower rivalry and its bipolar international system would unleash a new waveof weapons proliferation. Benjamin Frankel, "The Brooding Shadow: Systemic Incentives andNuclear Weapons Proliferation," The Proliferation Puzzle: Why Nuclear Weapons Spread andWhat Results, Zachary S. Davis and Benjamin Frankel, eds., (London: Frank Cass, 1993), p. 37.John J. Mearsheimer, "Back to the Future: Instability in Europe after the Cold War,"International Security, 19:3 (Winter 1994), pp. 5-49; Stephen Van Evera, "Primed for Peace:Europe After the Cold War," International Security, 15:3 (Winter 1990), pp. 7-57. See alsoBradley A. Thayer, “The Causes of Nuclear Proliferation and the Non-proliferation Regime,”Security Studies, 4:3, pp. 519.11 Several observers suggested that the Iraq case demonstrated that inspections, and byimplication the regime, are of limited value. Ruth Wedgewood, National Public Radio,“America and the World,” (8:00 pm ET), Transcript # 9513, April 1, 1995; Charles Duelfer,“The Inevitable Failure of Inspections in Iraq,” Arms Control Today, September 2002; Jason D.Ellis, “The Best Defense: Counterproliferation and U.S. National Security,” WashingtonQuarterly, 26:2 (Spring 2003), pp. 121; Charles Krauthammer, “Capitulation in Korea: Clinton'sCave-in Makes a Joke of the NPT,” Washington Post, January 7, 1994, p. A19.


p. 8

in the mid-1990s12, the 1998 South Asian nuclear tests13, the discovery of the Khan network, and

most recently with the controversy regarding Iran’s nuclear behaviour, analysts have solemnly

pronounced the demise of the regime and/or warned of a wave of new weapons states.14

Among the most vocal prognosticators of NPT collapse is the current US government. The

White House and its neoconservative theorists have long maintained that the treaty system is at

best obsolete and ineffective. They insist, therefore, on the need for a more “forward leaning” or

“aggressive” posture, one that includes counter proliferation, preventive war, and missile

defences.15

12 On North Korea, Robert McNamara refers to a “breakdown” of the regime, resulting in a waveof new weapons states such as Syria, Egypt, South Korea, Taiwan and Japan. For his part,conservative commentator Charles Krauthammer declared that North Korea meant that “NPT isdead.” Robert McNamara, “Meeting the Nuclear Challenge: A Personal Perspective,” April 25,2003,http://www.eisenhowerinstitute.org/programs/globalpartnerships/safeguarding/mcnamara1.htm;Charles Krauthammer, “Capitulation in Korea: Clinton's Cave-in Makes a Joke of the NPT,”Washington Post, January 7, 1994, p. A19.13 The South Asian tests, in particular, were accompanied by predictions of NPT collapse andnuclear proliferation. One former official, warned that “the world will begin to proceed downthe path toward widespread nuclear proliferation that we narrowly avoided back in the 1960s."Steve Coll, “The Race Is On -- and May Be Hard to Stop,” Washington Post, May 29, 1998, p.A01. See also Victor Galinsky, “Nuclear Proliferation After the Indian and Pak Tests,” inTwenty-First Century Weapons Proliferation: Are we Ready?, Henry Sokolski and James M.Ludes, eds., (London: Frank Cass, 2001), pp. 3-13; Marianne Hanson, “Future Disturbed byBlasts from the Past,” Courier Mail, June 1, 1998, p.13; John Mearsheimer, "Here We GoAgain," New York Times, May 17, 1998, p. 17; Thomas W. Lippman and John Ward Anderson,“U.S. Laws Curb Offers To Pakistan; Officials Look for Ways To Prevent Nuclear Test,”Washington Post, May 28, 1998, p. A25.14 Graham Allison, for example, has offered what can only be described as a vintage 1960sprediction of a proliferation debacle. Graham Allison, “The Cascade of Proliferation,” AtlanticMonthly. 296:3 (October 2005).15 On the Bush non-proliferation policy, see U.S, Department of Sate, “Interview: UnderSecretary John Bolton on U.S. Arms Control Policy,” August 14, 2001,http://www.usembassy.it/file2001_08/alia/a1081510.htm; John R. Bolton, “The BushAdministration's Forward Strategy for Non-proliferation,” Chicago Journal of InternationalLaw, 5:2, (Winter 2004), pp. 395-404; Jason D. Ellis, “The Best Defense: Counterproliferationand U.S. National Security,” Washington Quarterly, 26:2 (Spring 2003), pp. 115-133; SharadJoshi, “Unilateralism and Multilateralism: Analyzing American Nuclear Non-proliferationPolicy,” World Affairs, 167:4 (Spring 2005), pp. 147-161; Andrew Newman, “Arms Control,


p. 9

While this position represents the extreme end of non-proliferation thinking, much of the

underlying logic is widely accepted, i.e., that the diffusion of technological know-how and the

persistence of security threats make proliferation inevitable, and no treaty can stop it. This is the

same reasoning that provided the premise for 1960s predictions of widespread proliferation.

Even the Director General of the IAEA has reportedly declared that “30 countries could have

nuclear weapons within the next 10 to 20 years if efforts do not improve.”16

The main difference between today’s predictions and those half a century ago is that more

analysts are willing to acknowledge that the NPT has been useful, and that as a result, fewer

states than predicted have joined the nuclear club.17 A more sophisticated but nevertheless

pessimistic contemporary view concedes that the NPT has been successful in the past but

concludes that this previous success is irrelevant. 18 According to this school of thought, in a

post–Cold War, post–9/11 world of terrorists and “rogue states,” everything has changed.

Proliferation and Terrorism: The Bush Administration's Post-September 11 Security Strategy,”Journal of Strategic Studies, 27:1 (March 2004), pp. 59-88.16 The prediction is eerily similar to John F. Kennedy’s prediction. “El Baradei Warns OfNuclear Proliferation,” USA Today, December 14, 2005, p. 15. In his defence, the DirectorGeneral is conditioning the prediction on the premise that “efforts do not improve,” but that istantamount to saying that given the current system, the system will fail dramatically.17 This change may be the result of a small but growing number of scholars who have focused onnuclear restraint. See, for example, Mitchell Reiss, Without the Bomb: The Politics of NuclearNon-proliferation, (New York: Columbia University Press, 1988); Mitchell Reiss, BridledAmbition: Why Countries Constrain Their Nuclear Capabilities, (Washington: Woodrow WilsonCenter Press/Johns Hopkins University Press, 1995); Robert J. Einhorn, Mitchell B. Reiss, andKurt M. Campbell, eds., The Nuclear Tipping Point: Why States Reconsider Their NuclearChoices, (Washington, D.C. : Brookings Institution Press, 2004).18 One cannot help but recall William James’ three stages of a theory. “First, …a new theory isattacked as absurd; then it is admitted to be true, but obvious and insignificant; finally it is seento be so important that its adversaries claim that they themselves discovered it.” As regards thesuccess of the NPT, we are somewhere between stage one and stage two. William James,Pragmatism, (New York: Meridian, 1907), pp. 131-132.


p. 10

Indeed, what is striking about the conventional wisdom is its dogged persistence. In one form or

another, it has represented the consensus opinion for decades.19 As one observer noted, "almost

all the published predictions of the spread of nuclear weapons have been too pessimistic."20

B. Evidence and Arguments for Success and Failure

The overwhelming and historically consistent consensus of scholars and policy makers is that the

NPT and the regime has failed or is failing. But are they right? This section considers the

arguments from both sides.

B1. Evidence and arguments for success.

There are several types of arguments that support the view that the NPT has been a success. The

first type is statistical.

Statistical

One set of statistics looks at nuclear outcomes in the aggregate. In particular, one can point to a)

the declining rate of proliferation over time, b) the small percentage of countries that became

nuclear weapons states compared with the number of countries that considered doing so, and c)

the declining number of countries interested in acquiring nuclear weapons.

Proponents cite this record of restraint and make the additional point that these positive

developments follow or coincide with the establishment of the treaty. 19 See, for example, Matthew Woods, “Inventing Proliferation: The Creation and Preservation ofthe Inevitable Spread of Nuclear Weapons,” Review of International Affairs, 3:3 (Spring 2004),pp. 416-442.20 George H. Quester, "The Statistical “n” of “nth” Nuclear Weapons States," Journal of ConflictResolution, 27:1 (March 1983), p. 167. So far, I have found only two exceptions: Chafetz, whopredicts that pressures for proliferation will decrease in the years ahead and Donald Brennan,who at a conference in 1966, said that he would "bet even money that the numbers of nuclearpowers would not increase in the next ten to fifteen years." Glen Chafetz, "The End of the ColdWar and the Future of Nuclear Proliferation: An Alternative to the Neorealist Perspective," TheProliferation Puzzle: Why Nuclear Weapons Spread and What Results, Zachary S. Davis andBenjamin Frankel, eds., (London: Frank Cass, 1993), p. 128; Stanley Foundation, ProliferationUnlimited: A Strategy for Peace, [Report of a conference held March 10-11, 1966 inWashington, D.C.], (Muscatine, Iowa: Stanley foundation, 1966), p. 24.


p. 11

Consider, for example, the rate of proliferation. Measured as number of new nuclear weapons

states per decade, the rate of proliferation peaked in the 1960s and began to decline in the 1970s.

Perhaps not coincidentally, the NPT came into force in 1970.

Box 1. Rate of Proliferation, 1945-2000

(New Nuclear Weapons States Per Decade)

Proliferation Rate

0

0,5

1

1,5

2

2,5

3

3,5

1940s 1950s 1960s 1970s 1980s 1990s

Decade

Nu

mb

er

New

NW

S

Serie1

Sceptics could rightly point out that the chart does not include North Korea. Moreover, it might

be argued that the chart, while dramatic, is an artefact of small numbers. With so few cases, one

cannot be especially confident in the conclusions.

North Korea is not reflected in the chart, in part, because their nuclear status is unclear. Most

analysts believe that, consistent with North Korean claims, the Democratic People’s Republic of

Korea (DPRK) possesses at least one device. On the other hand, some long-time observers


p. 12

would not be surprised if it turned out that Pyongyang, in fact, had no nuclear weapons. More

importantly, the DPRK has repeatedly suggested that it is willing to renounce its arsenal, in

which case the corrected chart would nevertheless exhibit a declining rate of proliferation.

Indeed, as Matthew Bunn has observed, even if one includes the DPRK as a weapons state, there

are the same number of nuclear weapons states today as there were 15 years ago.21

It is certainly true that the chart reflects a small number of cases, but that, of course, is the point.

The very fact that there are a small number of cases suggests that the treaty has been successful.

Regardless, one has to use the data that is available, and it points to a thirty-year decline in the

rate of proliferation.

As discussed above, the number of countries that became nuclear weapons states is relatively

small, but evidence of nuclear restraint is not only found in this small number of states but in the

modest percentage of countries that acquired nuclear weapons. A much larger number of

countries considered, inherited, or acquired a nuclear option but maintained or reverted to a non-

nuclear status. Indeed, 75% of countries that could have become nuclear weapons states are

instead non-nuclear weapons states.

Box 2. Potential Nuclear Weapons States That

Remained Non-Nuclear

Argentina

Australia

Belarus

Indonesia

Iraq

Italy

S. Korea

S. Africa

Sweden

21 Personal communication with Matthew Bunn, December, 8, 2005.


p. 13

Brazil

Canada

Egypt

Germany

Greece

Japan

Kazakhstan

Libya

Norway

Romania

Switzerland

Taiwan

Turkey

Ukraine

Yugoslavia

A final statistical measure is the number of countries that aspire to become nuclear weapons

states. Contemporary analysts focus on North Korea and Iran, but how does that compare with

previous decades? There are, in fact, fewer states seeking nuclear weapons today than at any

point since WWII. The 1960s had the most nuclear aspirants. Indeed, the number of countries

that were interested in acquiring nuclear weapons in the 1950s and 1960s is roughly double the

total number of countries seeking nuclear weapons for the subsequent three decades combined.

As threatening as it may seem that a DPRK or Iran might seek to be nuclear weapons states,

policy makers from decades past found themselves in a far more threatening situation in terms of

proliferation.22 Simply put, since the NPT, fewer countries have had nuclear ambitions.

Behavioural

A second set of arguments draws on information about the behaviour of particular countries.

Take, for example, the behaviour of countries that wanted to keep a nuclear weapons option. In

general, these countries did not join the treaty immediately, despite the widespread belief at the

time that joining the treaty would make the import of nuclear technology substantially easier. In

other words, countries that had an interest in nuclear weapons did not view the treaty as just a bit 22 In context, it is difficult to imagine a more threatening possibility than the 1960s concern thatChina under Mao would get the bomb. Mao had said that nuclear weapons were paper tigers,that China could fight and survive a nuclear war, and that he would share nuclear weaponstechnology with other poor countries. In response, US policy makers explored a preventivenuclear attack in concert with the Soviet Union and considered sharing of US nuclear weaponswith India or Japan in order to balance against the Chinese bomb. On the perceived threat andAmerican policy options, see Francis J. Gavin, “Blasts from the Past: Proliferation Lessons fromthe 1960s,” International Security, 29:3 (Winter 2004), pp. 100-135; William Burr and Jeffrey T.


p. 14

of paper. They viewed it as a threat to a future nuclear weapons option and so avoided the treaty.

Some of these countries, such as India and Pakistan, went on to build nuclear devices. Others

remained outside the treaty until such time as subsequent governments were willing to renounce

their nuclear ambitions.

Consider the examples of Egypt and Australia, both of which repeatedly sought the acquisition

of nuclear weapons.23 Once joining the treaty, however, their nuclear acquisition activities went

to near zero.

The chart below maps nuclear decision-making in Australia from 1954 to 1998. The black bars

indicate the number of pro-acquisition decisions in a given year. The white bars reflect the

number of pro-renunciation decisions. “Decisions” consist of official governmental actions such

as commissioning a study on the feasibility of building nuclear weapons, asking an ally for the

transfer of nuclear weapons, or signing an arms control treaty. As the chart indicates, Australia’s

government made a number of pro-nuclear decisions for decades, but that came to an abrupt end

once the government ratified the NPT in 1973.

Richelson, “Whether to "Strangle the Baby in the Cradle: The United States and the ChineseNuclear Program, 1960-64,” International Security, 25:3 (Winter 2000-2001), pp. 54-99.23 Jim Walsh, Bombs Unbuilt: Power, Ideas, and Institutions in International Politics,(Cambridge, MA; MIT Doctoral Dissertation, May, 2000); Jim Walsh, "Surprise Down Under:The Secret History of Australia’s Nuclear Ambitions." Non-proliferation Review, 5:1 (Fall1997), pp. 1-20.

Box 3. Nuclear Decison Making in Australia, 1954-1998

2

1

1

2

3

4

08454 56 58 60 62 74 76 78 80 82 86 88 90 92 94 96

Pre-Regime

Decsions Favoring Acquisition

Decsions Favoring Renunciation

64 66 68 72700

Australia Joins Regime

A map of Egypt’s nuclear decision making between 1954 and 1998 reveals a similar pattern.

Prior to Egypt’s 1981 ratification of the NPT, Cairo made numerous pro-acquisition decisions.

Following ratification, that number goes nearly to zero. (The exceptions revolve around the

activities of Minister of Defence who was later sacked by President Mubarak.)

Box 3. Nuclear Decison Making in Egypt, 1954-1998

2

2

34

7 2

05 6 5 8 6 0 6 2 6 4 6 6 6 8 7 0 7 4 7 6 7 8 8 0 8 2 8 4 8 6 8 8 9 0 9 2 9 4 9 6 9 8

0

1

1

6

5 4

5Decisions FavoringAcquisition

Decisions FavoringRenunciation

Pre-Regime Egypt Joins Regime

This pattern can be observed in other states such as well Sweden, Germany, Italy, Canada, and

others. While not universal, the dynamic is strongly representative of state behaviour. Once

countries ratify the NPT, their pro-acquisition activities go to zero or decline precipitously. (The

important exceptions of Iraq, the DPRK, and Libya will be discussed in a later section.)

Discourse

A third type of argument points to contemporary discourse on proliferation. In the 1950s and

1960s, leading figures in the United States and elsewhere, viewed nuclear weapons as “normal,”

i.e., weapons that the militaries of most industrial states would incorporate into their forces. It

was expected that not only would countries acquire nuclear weapons but that tactical nuclear

weapons would be used in limited wars. Today, countries accused of seeking nuclear weapons

are called “rogue states.” In other words, they are not the norm; they are outliers—the

exceptions that prove the rule. The mental and normative conception of proliferation changed

dramatically in the decades following the NPT.

Institutional

A fourth set of arguments and evidence focuses on the regime itself, and in particular on various

institutional indicators regarding the health of the regime. There are several measures that

suggest that the NPT has gotten progressively stronger over time. To begin with, more and more

countries have joined the regime—so many that it now has almost the same number of members

as the United Nations itself. This is a claim that few international conventions can make.

It might be argued, however, that large membership may not mean much, if membership is easily

acquired and does not require that a country actually alter its behaviour. Even if one sets aside

the fact that renouncing the most powerful weapon in human history is not exactly a trivial

decision for many governments or that safeguards carry costs for member states, other aspects of

the NPT point to growing strength rather than weakness. First, the safeguards regime has gotten

progressively stronger over time—a process that has made it more intrusive, more demanding,

more costly, but also more effective. A second and related process has been the development of

new non-proliferation instrumentalities—from the interviewing of nuclear scientists to the use of

environmental sampling.

Here again, sceptics might respond that these innovations happened in response to regime

failures, e.g., full scope safeguards following the 1974 Indian test or the Additional Protocol

following the revelations about Iraq’s nuclear program in the early 1990s. It is a fair argument,

but in this case, the objection actually proves the point: the system’s response to crisis was the

development of new strength rather than collapse. Damage followed by repair and growth is

usually associated with robust systems. Failing systems, in fact, do not respond to challenges

and therefore grow weaker and die.

A final indicator of the regime’s increasing rather than declining strength is the growing political

significance of the International Atomic Energy Agency (IAEA). Today, if the IAEA issues a

finding that a country is violating its safeguards agreement, it has major political and policy

consequences. It is a result that most states will vigorously seek to avoid. It is also a result that

the United Nations Security Council and the major powers will find difficult to ignore even if

they prefer to do so. Granted, in the Iraq-case the IAEA could not stop a superpower from going


p. 18

to war in the name of non-proliferation—a result no international organization or country could

have accomplished—but the agency was strong enough to issue a finding that ran contrary to the

desires of the world’s only superpower. Indeed, it is difficult to name many, if any, international

organizations that have the same power to affect national policy as the IAEA.

Taken together, these institutional measures of health and weakness are strikingly inconsistent

with the portrait of the NPT as failing or near collapse. If the treaty system were failing, one

would not expect more countries to be joining the treaty, stronger and evolving inspection

instrumentalities, and increasing political influence for the agency that implements the treaty. If

one adds the institutional indicators to the statistical, behavioural, and discourse trends, the result

is a very powerful and consistent picture of a regime that has had a remarkable effect on nuclear

outcomes.

B2. Evidence and arguments for failure.

Where some see success, others see failure. Critics of the NPT’s performance tend to cite at

least one of four arguments. The first is the continued existence of nuclear-weapons’ states in

and outside the treaty. A second concerns cheating, and in particular the cases of Iraq, the

DPRK, Libya, and Iran. The third is that the treaty is ill equipped to handle the threats posed by

non-state actors, especially terrorists and black marketeers. The final critique concerns the treaty

itself, namely, its lack of enforcement provisions, its withdrawal clause, and its “loopholes”

regarding bomb-relevant fuel cycle technology.

Continued Existence of Nuclear Weapons States

It is certainly true that nuclear weapons states continue to exist. Of course, the five “official”

nuclear weapons states acquired these weapons before the treaty and the other three unofficial

nuclear weapons states did not join the treaty. Most observers interpret Article VI as obliging

the five NPT nuclear-weapon states to eliminate their nuclear stockpiles, and even the most

fervent treaty supporter would have to concede that is unlikely for the foreseeable future. Even

though weapon stockpiles in the United States and Russia have declined in recent years, these

reductions seem unrelated to the states’ NPT commitments. Indeed, there are signs that some


p. 19

elements within the weapons states wish to abandon this obligation or give even greater political

and military priority to their nuclear arsenals. In short, the NPT has not successfully resulted in

disarmament.

Cheating by NPT Member States

One also cannot ignore the record of cheating on the part of some NPT members. While most

countries did not use their treaty membership as a vehicle for support of a covert weapons

program (see above), it is apparent that Iraq and the DPRK did. Moreover, it is clear that Iran

engaged in a broad program of concealment in violation of its safeguards obligations and that

Libya also acted in direct contravention to its safeguards commitments.

Still, the issue of cheating is not uncomplicated. As the Iraqi Survey Group reported, it was fear

of NPT and Security Council non-proliferation inspections that led the Hussein government to

abandon its nuclear program in the early 1990s. It was IAEA inspectors who, in the course of

carrying out their safeguards responsibilities, discovered the discrepancies in the North Korean

declarations. (Indeed, North Korea has announced that it is willing to renounce its nuclear

arsenal and wants to rejoin the NPT and submit to IAEA inspection.)

Iran and Libya present somewhat different cases. Some observers of the Iranian program believe

that Tehran seeks a nuclear-weapon capability but does not want to openly violate the NPT and

become another North Korea. Instead, it is seeking to provide “objective guarantees” concerning

its program and plans to act in accordance with the letter, if not the spirit, of the law. Libya’s

activities were not discovered by the IAEA, nor was the renunciation of the program engineered

by agency officials. Nevertheless, it has renounced its program and is seeking to regain its status

as a treaty member in good standing. (It is also unclear whether, in the absence of British

intervention, Libya would have become a nuclear-weapon state even with its procurement

program.)

Indeed, with the (likely) exception of the DPRK, none of the cheaters have actually become

nuclear-weapon states, and the DPRK says that it hopes to give up its program. Moreover, the


p. 20

legal basis for discovering and acting against these programs has been the treaty itself. What is

striking, in fact, is that despite these transgressions (and the South Asian nuclear tests), there has

been no rush to nuclear weapons. No country is anxious to become the next North Korea, and no

country has spoken out in defence of North Korea’s violations.

Terrorism and Black Markets

Regime critics are right to suggest that the NPT was not designed to address the threats posed by

terrorists or nuclear black markets. Still, it has to be said that a larger number of nuclear

weapons states would increase the risk of both nuclear terrorists (who might receive or steal

material from these states) and black marketeers (who might make use of a state program in the

same way that the Khan benefited from the Pakistani program). Indeed, it is hard to imagine that

a similar network that was wholly unconnected to a weapon state would have been as successful.

It is also worth noting that the basic defence against nuclear terrorism—securing nuclear

material—is a logical extension of material accounting and other practices associated both with

safeguards and with other, related parts of the regime, e.g., the Convention on the Physical

Protection of Nuclear Material and export control efforts. This is not to suggest that current

efforts are up to the task. They are not. Rather, the relevant fact here is that the NPT has

developed institutions, instrumentalities, and political understandings that can help facilitate

policies that are required to address these challenges.

Treaty Loopholes

A final complaint about the NPT and an alleged reason for its failure relates to the treaty’s

provisions—both what they include and what they do not include. There is, for example, no

enforcement provision. Article X permits withdrawal. The treaty, by custom, also permits states

to acquire enrichment and reprocessing technologies that could enable a short sprint to the bomb.

The lack of enforcement provisions has been a topic of particular interest to American analysts,

where it is often seen as a critical flaw—in the words of one study, an “Achilles heel.”

Invariably, these critics want to see a more muscular regime and, in some cases, they want


p. 21

enforcement mechanisms that approach automaticity. There are numerous reasons to doubt the

wisdom of “automatic” solutions, and proponents of such systems fail to explain how such

measures would be adopted by a sceptical international community. What is most relevant,

however, is the fact that the NPT has achieved the success described above without an explicit

enforcement clause.

Have transgressors gone unpunished? Hardly. Iraq and North Korea have been subjected to

political penalties, economic sanctions, and the threatened or actual use of force. (The case of

Iran is comparatively recent and still in the making.) The absence of an explicit enforcement

clause has meant that enforcement has fallen to the United Nations or bilateral responses by

individual states. While lacking in uniformity, this structure has meant that different remedies

have been tailored for different cases.

Finally, even robust enforcement provisions would not enable the IAEA to act against states

outside the treaty, e.g., the Indian and Pakistani nuclear programs. The UN and individual states

will continue to have to play an active role in enforcement even if the NPT had an enforcement

provision. Indeed, as all students of international organizations recognize, the whole notion of

IAEA enforcing rules independently of its member states is naïve. The agency is only as strong

as the political will of its members, and it is those members who already possess the

responsibility for enforcement.

Another concern about the treaty is Article X, the withdrawal provision. North Korea’s

withdrawal and the possibility of Iran’s withdrawal from the NPT have spurred questions about

the adequacy of this provision. Some member states want to make it more difficult to withdraw.

Others have suggested that withdrawal should carry certain obligations such as the return of

nuclear technology acquired during NPT membership. This is a useful conversation and may

result in ideas that strengthen the regime. Still, it would be a mistake to consider the current

withdrawal language a fatal flaw.


p. 22

From a historical perspective, the withdrawal provision served a very useful purpose. At the

dawn of the regime, joining the NPT was not uncontroversial. There were heated debates in

many capitals about the wisdom of signing the NPT. Treaty opponents cited a long list of

reasons why the treaty would have unwelcome consequences, everything from the need to keep a

nuclear option to concern about foreign spies (qua agency inspectors) running around stealing

economic and military secrets. Treaty proponents had a powerful answer: even if these concerns

were valid, the country could always exercise its Article X rights and withdraw from the treaty.

The withdrawal provision took the sting out of many of the opponents’ arguments and helped

pave the way for NPT membership. Today, membership is less of an issue, given the near

universal subscription to the treaty, and so a discussion about withdrawal is probably

appropriate.

Nevertheless, the issue should be kept in perspective. Despite the passage of over half a century,

only one state has exercised the withdrawal option—and now that state says it wants to rejoin the

treaty.

By far the most serious concern about the NPT relates to the concept of peaceful uses and the

ability of states to acquire fuel cycle technologies and achieve a virtual nuclear weapons

capability. This issue has always been a concern, but it has reached particular prominence in

recent years given the case of Iran and its undeclared procurement and development activities.

Many, including the Director General of the IAEA and thePresident of the United States, have

called for changes in fuel cycle policy.

The spread of enrichment and reprocessing technologies is, in fact, a major threat to non-

proliferation and to the regime. The treaty had a similar problem with respect to Article V and

peaceful nuclear explosives (PNE’s). Though the treaty guaranteed a nation the right to PNE’s,

states parties arrived at an unwritten political consensus that this right would not be exercised. A

similar remedy may be required here, but it will be much more difficult under current conditions

and, in any case, achieving a similar consensus will take time. Unfortunately, the same is also

true for the other proposed remedies to this problem, such as international or multilateral


p. 23

ownership of enrichment and reprocessing or a top down denial of technology by advanced

industrial states.

In sum, there are legitimate complaints and concerns about the treaty, but many of the reasons

cited for why the NPT is fatally flawed are overdrawn. The problems of enforcement and

withdrawal, while worthy of discussion and new thinking, have not fundamentally impeded the

success of the treaty. The same is true of the issue of the fuel cycle, but that provides little

comfort. Of all the worries about the treaty, the issue of the fuel cycle is the most dangerous and

perhaps the most difficult to resolve in the current international political climate.

C. Measuring Success and Failure

After reviewing the arguments for success and failure, it is easy to see merit in the positions of

both sides. Overall, the treaty appears to have restrained the pace of proliferation to an extent

unimaginable by scholars, policy makers, or even the diplomats who negotiated the treaty. On

the other hand, the weapons states have not honoured their Article VI obligations, states within

the treaty have cheated, and access to fuel cycle technology may subvert the core objectives of

the treaty. So what are we to make of these crosscutting results?

First, one must ask, “What is the basis for judging success?” Success or failure cannot be

measured in the abstract. It must be defined by some standard.

Most NPT critics appear to believe that perfection is the appropriate standard. By this measure,

the identification of a single cheater or one flaw qualifies the treaty as a failure.

A second basis for measurement might be called the “pragmatists’ standard.” What is important

here is not perfection, but rather how well something performs compared to how it could have

reasonably been expected to perform. In short, did the NPT beat expectations? Alternatively,

one could compare it with the other alternatives that were available at the time. This requires

counterfactual reasoning, which can be difficult. Yet another variant of the pragmatic standard


p. 24

also employs counterfactual reasoning. It asks the evaluator to imagine a world without the NPT

and then to consider whether the world would have been better off or not without the treaty.

A third standard, one typically favoured by social scientists, is performance compared to other

domains. So for example, how does NPT performance compare with other treaties in general or

other arms control treaties in particular?

No standard is without its flaws. The perfection standard is problematic on its face, given the

absence of perfection in any human endeavour. Moreover, the perfection standard would seem

most likely to the mistakes associated with a failure to understand success.

The pragmatist and “comparison with comparable phenomena” standards are also not without

problems. Beating expectations may be too low a standard, and the utility of the “comparable

phenomena” standard depends on the quality of the referents. For some events or experiences,

there are few comparable phenomena, and comparisons on the wrong dimensions may be more

misleading than helpful. Both the pragmatism and comparison standards also fail to give full

weight to the objectives of the performance being evaluated. Given a goal of preventing nuclear

proliferation or ultimately, nuclear use, beating expectations or performing better than other

treaties may be insufficient given the importance of the objective and the significance of the

exceptions.

As there is no one ideal standard of evaluation, what can be said about the NPT? Obviously, like

everything else in life, it fails the perfection standard, but it does quite well when judged by the

other standards. It certainly beat expectations: virtually every policy maker and analyst since

1945 has wrongly predicted the future of proliferation by overestimating the degree to which

nuclear weapons would spread. In addition, no one predicted that states would give up their

nuclear weapons or that the interest in nuclear weapons would decline over time. Moreover, it is

hard to imagine that the world would have fared better without the NPT or that other

contemporary alternatives (e.g., selective proliferation to China and Japan or the Multilateral

Force) would have generated a superior outcome.


p. 25

Even when considering that the goal of stopping nuclear proliferation carries higher than

average stakes and has a low tolerance for exceptions, the NPT does well. Judged by outcomes,

no NPT state with the possible exception of the DPRK has acquired nuclear weapons. It is

difficult to think of a similar level of success enjoyed in any domain related to national

security—or in policy in general for that matter. This is not to minimize the challenges,

particularly relating to the fuel cycle. Moreover, as television ads for investment companies

warn, “past success is not a guarantee of future performance,” but it would seem difficult to

ignore the striking success of the NPT.

D. Why Is the NPT’s Success Ignored?

The NPT’s contribution to nuclear restraint is arguably one of the 20th century’s biggest—and

unheralded—public policy achievements. This success raises a number of questions. First and

foremost, it raises the question of causes. What accounts for the treaty’s success? That

fundamental question is addressed in the next section. Here the focus is instead on the apparent

disconnection between performance and perception. Why do policy makers and scholars

continue to emphasize failure in the face of unprecedented and unexpected success?

The focus on failure is pervasive and consistent over time, i.e., it is systematic. Are there reasons

that explain this bias? What does the bias tell us about ourselves and our theories of

international behaviour?

Explaining the systematic bias on the part of policy makers would appear to be fairly

straightforward. Decision makers understandably tend to concentrate their attention on present

and near-term dangers, not on comparing relative risk over time. And as with the media, there is

a natural preoccupation with danger rather than success. In the triage of policymaking, a danger

(something not working) will receive greater attention than a success (something that is

working).


p. 26

More speculatively, it might be said that inevitability arguments (e.g., it is inevitable that nuclear

weapons will spread, it is inevitable that country X will acquire nuclear weapons) are especially

seductive. It may be that the strong premises (e.g., capability will spread over time, countries

face security threats) combined with a simplifying frame (inevitability) are hard to resist.

Issues of political self-interest may also contribute to a pre-occupation with faults and failures.

Regime critics, who would prefer that states give increasing attention to non-regime based policy

instruments such as defences and preventive war, are naturally going to challenge the value of

the treaty. In the run up to the 2003 war in Iraq, for example, American officials repeatedly

questioned the effectiveness of the regime as part of their public case for preventive war.

Paradoxically, regime advocates indulge in similarly negative rhetoric. In order to muster

political support for the treaty, they claim that the NPT is in danger of collapse or that a crisis is

at hand. Thus, both critics and proponents have their own reasons for emphasizing failures

rather than successes.

The apparent bias of scholars is more difficult to explain. Part of the reason may be related to

the fact that for most of the nuclear age, the focus in security studies has been on the U.S., the

Soviet Union and to a lesser extent on the other nuclear weapons states. In other words, the

primary cases in the field are all countries that acquired nuclear weapons (non-proliferation

failures), not countries that abstained from nuclear weapons (non-proliferation successes). In the

last ten years, there has been more attention paid to the nuclear histories of other countries and

somewhat more attention to regime successes, but the change has been limited.

Alternatively, it may have something to do with the evolution of the field of international

relations. Scholars drawn to the Realist school of international relations may be more likely than

others to pursue work in security studies. They, in turn, come to the topic with a firmly

entrenched scepticism about the importance of international institutions. Looked at from a

different angle, it may be that Institutionalists tend to go into the field of political economy rather

than security studies. The post-modernists, for their part, have not had much to say about the

issue.


p. 27

Of course, these are just musings and guesses, but the disconnect is sufficiently wide and the

topic sufficiently important that it is probably worth pursuing the question. An answer might tell

us a lot about ourselves as analysts and help us avoid similar errors in the future.

IV. What Accounts for the Success of the NPT Regime?

With the success of the NPT being largely ignored, it is not surprising that the causes of that

success have also received short shrift. Regime sceptics have contended that the NPT is just a

piece of paper, subject to the whims of great powers and vulnerable to cheaters. Regime

advocates, while not offering a clearly stated theory of success, implicitly suggest that the power

of the NPT is the power of international law and international public opinion.

In this section, these claims are reviewed, and both are found wanting. Critics fail to appreciate

the constraining effect of the NPT, and advocates provide a narrow explanation for a dynamic

that goes well beyond a respect of the law. It will be argued that the power behind the NPT is

not legal but rather institutional and political.

Before explaining why the NPT has enjoyed unexpected and little appreciated success, it is

useful to first step back and consider the broader context of nuclear restraint. This restraint

presents somewhat of a puzzle. In solving this puzzle, one can better understand how the NPT

achieved its goals.

A. The Puzzle

A major mystery confronting any student of international affairs is that virtually every indicator

and historical development since 1970 would suggest that nuclear weapons should have spread

faster than they have.

Consider the period between the birth of the NPT and the collapse of the Soviet Union. During

these two decades, civilian nuclear technology spread, and thus the technical barriers to


p. 28

acquisition became progressively lower. States with nuclear weapons paid few if any costs for

their status as nuclear weapons states. (The possible exception might be India, which endured

modest but relatively short-lived penalties following its 1974 test.) If anything, nuclear-weapon

states enjoyed or were perceived to enjoy benefits from their nuclear weapons status, including

everything from increased prestige to enhanced political power (e.g., permanent seats on the UN

Security Council). Moreover, as was discussed in previous sections, the NPT seemed to suffer

from a number of important flaws: the absence of an enforcement provision, weak verification,

and a withdrawal clause that would permit countries to acquire nuclear technology and then

renounce their obligations. Most important of all, the treaty did not address the underlying

security motivations that were and are widely believed to fuel the spread of nuclear weapons.

The puzzle, plainly put, is thus the following: How could a treaty with no enforcement, weak

verification, and no ability to remove or resolve security threats be so successful?

Now consider the 15 years since the end of the Soviet Union. If anything, the puzzle becomes

even stronger. All of the previously listed problems persisted (spreading technology, flaws in the

treaty, etc.), but new pressures have accrued that should have portended greater proliferation.

Following the end of the Cold War, several scholars of international relations predicted that the

end of the stable bipolar international system would bring with it a rash of proliferation. These

predictions were made even before it became apparent that there were hundreds of tons of

unsecured fissile material in the former Soviet Union.

This period also witnessed the first Gulf War, the lesson of which—according to a leading Indian

official—was that the only way a country could protect itself from the United States was through

the acquisition of nuclear weapons. The post-1990 period was also marked by the well-

publicized transgressions in Iraq and North Korea, the South Asian atomic tests, and the

discovery of a nuclear black market based in Pakistan.


p. 29

If one had told the framers of the NPT that their treaty would have to survive in such a hostile

environment, even the optimists might have blanched. How could the treaty survive, let alone

succeed, in such a world?

B. Explaining the Puzzle: Potential Explanations

To explain the puzzle of success, one could turn to a number of traditional explanations. Several

of the explanations look beyond the NPT and the regime to account for nuclear restraint. It has

been suggested, for example, that superpower alliances like NATO and the Warsaw Pact

curtailed the spread of nuclear weapons.24 Others point to superpower pressure, i.e., threats by

the superpowers that persuaded or compelled states to abandon their nuclear ambitions. Still

others point to the role of supply-side constraints.25 The most common regime-related

explanations tend to emphasize two factors: a) the salutary effects of a growing international

24 On the positive effects of security guarantees, see Frankel, "The Brooding Shadow: SystemicIncentives and Nuclear Weapons Proliferation," pp. 37, 46; Kenneth N. Waltz, "The Spread ofNuclear Weapons More May Be Better," Adelphi Papers, No. 171, (London: IISS), Autumn,1981, pp. 78-79; Kenneth N. Waltz, "More May Be Better," The Spread of Nuclear Weapons,Scott D. Sagan and Kenneth N. Waltz, eds., (New York: W.W. Norton & Company, 1995), p.42; Stephen Van Evera, "Hypotheses of the Causes of the Rate of Nuclear Proliferation,"Unpublished paper, March 4, 1976, p. 6; Thayer, “The Causes of Nuclear Proliferation and theNon-proliferation Regime,” p. 503; see also Greenwood, Potter, and Meyer in Stephen M.Meyer, The Dynamics of Nuclear Proliferation, (Chicago: University of Chicago, 1984), pp. 68,102.25 On denial as a non-proliferation tool, see Peter Clausen, "U.S. Nuclear Exports and the Non-proliferation Regime," in Limiting Nuclear Proliferation, Jed C. Snyder Samuel F. Wells, eds.,(Cambridge: Ballinger, 1985), pp. 183-212; Victor Galinsky, "Restraining the Spread of NuclearWeapons: A Walk on the Supply Side," in Limiting Nuclear Proliferation, Jed C. Snyder SamuelF. Wells, eds., (Cambridge: Ballinger, 1985), pp. 255-283; C. Raja Mohan, "Why Nations GoNuclear: An Alternate History," in Nuclear Proliferation in the 1980s: Perspectives andProposals , William Kincade and Christopher Bertram, eds., (New York: St. Martin's Press,1982), p. 38; William Potter, "Soviet Nuclear Export Policy, in Limiting Nuclear Proliferation,"in Limiting Nuclear Proliferation, Jed C. Snyder Samuel F. Wells, eds., (Cambridge: Ballinger,1985), pp. 213-252. For a critique of denial approach, see George Rathjens, "RethinkingNuclear Proliferation," in Weapons Proliferation in the 1990s, Brad Roberts, ed., (Cambridge:MIT Press, 1995), pp. 93-106.


p. 30

norm against the possession of nuclear weapons26 or b) the moral or political force of

international law.27

As with most things in life, it is doubtful that nuclear constraint can be attributed to just one

cause or explanation. Nevertheless, these candidate explanations do not appear to be very

promising. A full explication of their flaws requires a separate essay, so a brief survey is offered

instead.

B1. The alliance explanation. 26 On the taboo against the use of nuclear weapons, see Nina Tannenwald, "The Nuclear Taboo:The United States and the Normative Basis of Nuclear Non-Use," International Organization,53:3 (Summer 1999); Richard Price and Nina Tannenwald, "Norms and Deterrence: The Nuclearand Chemical Weapons Taboos," in The Culture of National Security: Norms and Identity inWorld Politics, Peter J. Katzenstein, ed., (New York: Columbia University Press, 1996), pp. 114-152; Nina Tannenwald, “Stigmatizing the Bomb: Origins of the Nuclear Taboo,” InternationalSecurity, 29:4 (Spring 2005), pp. 5-49. Nehru's resistance to developing an Indian bomb andJapan's "nuclear allergy" are sometimes cites as examples of norm-driven behavior. On theNehru, see Sumit Ganguly, "India's Pathway to Pokhran II: The Prospects and Sources of NewDelhi's Nuclear Weapons Program," International Security, 23:4 (Spring 1999), pp. 150-151;Stephen M. Meyer, "Nuclear Decision Making in India," (Cambridge: Center for InternationalStudies, MIT, 1981), pp. 5, 9. For a different view of Nehru, see George Perkovich, India'sNuclear Bomb, (Berkeley: University of California Press, 1999). On Japan, see, for example,Peter J. Katzenstein and Nobui Okawara, Japan's National Security: Structures, Norms, andPolices, (Ithaca: East Asia Program, Cornell University, 1993), pp. 165-171; Selig S. Harrison,"Japan and Nuclear Weapons," in Japan's Nuclear Future, Selig S. Harrison, ed., (Washington:Carnegie Endowment for International Peace, 1996), pp. 5, 14-18. Reiss, for example,maintains that norms -- what he calls "the general consensus against nuclear weapons" -- havebeen a contributing but secondary factor in nuclear restraint, particularly in the cases such asSweden. Dunn credits anti-nuclear norms with anti-nuclear outcomes in "Indonesia, thePhilippines, Singapore, Venezuela, Mexico, Chile, Egypt, Algeria, Nigeria, and Yugoslavia."Sagan argues that norms help explain why Ukraine gave up its nuclear weapons. Mitchell Reiss,Without the Bomb: The Politics of Nuclear Non-proliferation, pp. 263-265; Lewis A. Dunn,"Four Decades of Nuclear Non-proliferation: Some Lessons from Wins, Losses, and Draws,"Washington Quarterly, 14 (Summer 1999), p. 7; Scott Sagan, "Why Do States Build NuclearWeapons? Three Models in Search of a Bomb," International Security, 21:3 (Winter, 1996-1997), pp. 80-82.27 On treaty compliance, see, for example, Abraham Chayes and Antonia Chayes, "OnCompliance," International Organization, 47:2 (Spring 1993), pp. 175-205; R. B. Mitchell,


p. 31

The alliance explanation suffers from the fact that many alliance members sought nuclear

weapons despite the presence of an alliance (e.g., the United Kingdom, France, Australia,

Germany, Italy, China,) while other countries remained non-nuclear despite their lack of an

alliance.28 Moreover, at various points in U.S. history, American presidents looked favourably

on the idea of a selective spread of nuclear weapons to allies. This was certainly true during the

Eisenhower administration, which developed the concept of “nuclear sharing.” It was also true,

though to a lesser extent, during the Nixon administration.

Looking back, one could even argue that nuclear alliances produce a countervailing contagion

effect, i.e., being in an alliance with a nuclear power actually increases interest in acquiring

nuclear weapons so that one can have equal or advantageous footing vis-à-vis the alliance

hegemon and the other members of the alliance (e.g., the United Kingdom, Australia). In any

case, the power of such alliances should have weakened after the fall of the Soviet Union (the

point made by the international relations scholars who predicted a post–Cold War wave of

proliferation), but a decade and a half later, this effect appears not to have taken place.

B2. The superpower pressure explanation.

The superpower pressure explanation is no less problematic. Obviously, the United States did

not sufficiently pressure or punish France, the United Kingdom, Israel and other states that did

acquire nuclear weapons. It is also unlikely, given the Cold War competition, that the United

States would have kicked an important ally out of the alliance or crippled it with sanctions—an

outcome that could only have been welcomed by the Kremlin.29

“Regime Design Matters - International Oil Pollution and Treaty Compliance,” InternationalOrganization, 48: 3 (Summer 1994), pp. 425-458.28 One might also include India and North Korea, which went forward with nuclear programsdespite treaties with the former Soviet Union.29 On the very day the President Nixon announced he would seek ratification of the NPT, heissued a National Security Decision Memorandum directing that American allies -- particularly,the West Germans -- not to be harassed over the treaty. What mattered in Nixon’s realpolitikwas the Soviet competition and the alliance system, not international treaties. Seymour M.Hersh, The Price of Power: Kissinger in the Nixon White House, (New York: Summit, 1983), p.148; Seymour M. Hersh, Samson Option: Israel, America and the Bomb, (London: Faber, 1993),


p. 32

In specific cases, the pressure thesis has actually been disproved by the archival documents.30

More generally, the pressure thesis is inconsistent with the record of NPT ratification. Many of

the most important states did not ratify the NPT until the mid-1970s and 1980s—long after the

treaty had opened. The United States either failed in its attempt to muscle states into the treaty

or did not try. Given the foreign policy team in charge at the time, the latter is the most likely

explanation.

This is not to suggest that pressure was never tried and never successful. There is a widespread

belief that United States pressure did constrain nuclear programs in South Korea and Taiwan,

though it must be said that neither country's nuclear history has been subject to systematic

scholarly inquiry.31 As a general explanation, however, the pressure thesis has logical and

empirical problems.

B3. The supply-side controls explanation.

Supply-side controls such as bilateral export controls and the efforts of the Nuclear Suppliers

Group offer something of a paradox. Most analysts and policy makers list supply-side controls

pp. 209-210; Glenn T. Seaborg, Stemming the Tide, (Lexington, MA: Lexington, 1987), pp. 53-54.30 For example, American officials specifically stated they would avoid pressuring Australia overits nuclear program. National Archives II (NAII), Memo for the President, Subject: YourMeeting with the Prime Minister of Australia, April 29, 1969 (Secret), Memo for signatureattached to memo from John P. Walsh, Acting Executive Secretary, to the Secretary of State, p.1, Department of State Central Files, 1967-1969, Folder Pol 7, 3/1/69, Australia, Box 1842;NAII, Enclosure: Additional Talking Points, attached to Memo for the President, Subject: YourMeeting with the Prime Minister of Australia, April 29, 1969 (Secret), pp. 4-5.31 On the South Korean and Taiwanese nuclear programs see Michael J. Mazarr, North Koreaand the Bomb, (London: Macmillan, 1995), pp. 25-28; Don Oberdorfer, The Two Koreas, (NewYork: Basic Books, 1997), pp. 68-74; Reiss, Without the Bomb: The Politics of Nuclear Non-proliferation, pp. 78-108; Ta-you Wu, "A Footnote to the History of Our Country's 'NuclearEnergy' Policies," Biographical Literature, May, 1988, Translation from Chinese by ISIS;http://www.isis-online.org/publications/ index.html; Joseph A. Yager, "Northeast Asia," Non-proliferation and U.S. Foreign Policy, Joseph A. Yager, ed., (Washington: Brookings Institution,1980), pp. 47-81; Frankel, "The Brooding Shadow: Systemic Incentives and Nuclear WeaponsProliferation," pp. 48-51.


p. 33

among the most important non-proliferation policy tools and yet virtually everyone agrees that

dedicated proliferators can sidestep the obstacles they pose and acquire nuclear technology.

Whether this is true or not is difficult to say, but one can point to cases (India in the 1970s, Iraq

in the 1980s, North Korea in the 1990s) that seem to suggest that the system is rather porous. It

is certainly hard to argue that supply-side controls offer a general explanation given a context in

which nuclear technology and materials are increasingly available and when developing

countries from Pakistan to the DPRK to Brazil and Iran are able to acquire at least elements of

enrichment and reprocessing.

B4. The anti-nuclear norm explanation.

It is probably true that more people today subscribe to the idea that nuclear weapons are

illegitimate than in previous decades. Of course the exceptions are also noteworthy, and they

include large numbers of citizens in the official nuclear weapons states as well as in India and

Pakistan, where nuclear weapons have become a powerful symbol that is often used to win

popular support. Countries driven by a desire for recognition, prestige, or national status appear

especially resistant to the anti-nuclear norm.

The major problem with the norms explanation, however, is not that the norm lacks universality

or that it is a rather recent phenomenon (and thus cannot explain nuclear restraint for at least the

first half of the nuclear age). The main problem with the norms explanation is that most

decisions to pursue nuclear weapons are not made in public. They are almost always secret.

Indeed, they are among the most secret activities that a state pursues. The public does not decide

nuclear policy; it finds out about nuclear decisions after the fact. Moreover public opinion only

matters if the country in question is democratic and even then, it is not axiomatic that opposition

will translate into a change in policy.

B5. The international law explanation.

The international law explanation is most often offered by diplomats, lawyers, and legal

scholars.32 It suggests that compliance with the NPT—despite the treaty’s flaws and despite a

variety of external pressures—is a consequence of states’ respect for international law.

Governments feel a moral commitment or see it in their self-interest to abide by their

international commitments.33 As a moral or legal explanation, the invocation of international

law raises a number of questions.34 For example, state compliance with international law is

uneven at best. Why does moral force or respect of the law apply to the NPT but not other

treaties? In fact, scholars of international relations have suggested that regimes work best in

32 Among political scientists those who cite the salutary effects of non-proliferation regime areRosecrance, Greenwood, Potter, Meyer, Quester, Smith, Reiss, Mandelbaum, and Sagan. Inaddition, both Chafetz and Solingen make regime-related arguments. On Rosecrance,Greenwood, Potter, Meyer, see Meyer, The Dynamics of Nuclear Non-proliferation, pp. 68, 102;see also Quester, "The Epistemology of Nuclear Proliferation," Journal of International Affairs,40 (Summer, 1986), p. 178; Reiss, Without the Bomb, pp. 260-263 Michael Mandelbaum,"Lessons of the Next Nuclear War," Foreign Affairs, 72:2 (March-April 1995), p. 24; Sagan,"Why Do States Build Nuclear Weapons? Three Models in Search of a Bomb," pp. 76, 80-82;Glenn Chafetz, "The Political Psychology of the Nuclear Non-proliferation Regime," TheJournal of Politics, 57:3 (August 1995); Etel Solingen, "The Political Economy of NuclearRestraint," International Security, 39:2 (Fall 1994).33 On regimes in general, see Stephen Krasner, ed., International Regimes, (Ithaca: CornellUniversity Press, 1983); Robert O. Keohane, After Hegemony: Cooperation and Discord in theWorld Political Economy (Princeton: Princeton University Press, 1984); Robert Keohane,“Neoliberal Institutionalism: A Perspective on World Politics,” in International Institutions andState Power, Robert Keohane, ed., (Boulder: Westview, 1989); Oran R. Young, "InternationalRegimes: Problems of Conception Formation," World Politics, (April 1980), pp. 331-356; OranR. Young, "International Regimes: Toward a New Theory of Institutions," World Politics, 39:1(1986), pp. 104-122; Stephen Haggard and Beth A. Simmons, "Theories of InternationalRegimes," International Organization, 41:3 (1987), pp. 491-517; Lisa L. Martin and Beth A.Simmons, "Theories and Empirical Studies of International Institutions," InternationalOrganization, 52:4, pp. 729-757.34 Scholars have identified other reasons that might explain the influence of regimes. Moravcsikmaintains that states are drawn to regimes, because they dampen domestic political uncertainty.Barnett and Finnemore focus on the ability of international organizations to make rules andcreate social knowledge. Andrew Moravcsik, "The Origins of Human Rights Regimes:Democratic Delegation in Postwar Europe," International Organization, 54:2, pp. 217-252;Michael N. Barnett and Martha Finnemore, "The Politics, Power, and Pathology of InternationalOrganizations," International Organization, 53:4 (Autumn 1999), pp. 699-732; MarthaFinnemore, "International Organizations as Teachers of Norms: The United Nations Educational,Scientific, and Cultural Organization and Science Policy," International Organization, 47:4(Autumn 1993), pp. 565-598.


p. 35

domains other than those having to do with security.35 When it comes to issues of security, a

state is seen as more likely to disregard moral and legal claims in favour or self-interest and

survival.

C. Conclusion: A Missing Element.

To be clear, none of the objections raised against these explanations is fatal. No doubt all of

them have contributed to nuclear restraint by some governments at particular moments in time.

Nevertheless, they are all sufficiently limited that they do not amount to a compelling resolution

of the puzzle, even if one adds them all together. Put another way, the explanations do not seem

as strong as the puzzle.

This is a problem for both regime sceptics and regime advocates. If both sides acknowledge that

the restraint of proliferation has been far greater and far more effective than anyone could have

imagined, then this success must be explained. The problem is that neither non-regime based

explanations favoured by regime sceptics (e.g., alliances and export controls) or the standard

explanations favoured by regime advocates (norms, respect for international law) seem up to the

task.

Something must be missing. In order to try to find out what it is, the next section looks to the

process of nuclear decision-making. What does nuclear decision-making as a policy or political

process, tell us about nuclear restraint?

V. The Primacy of Politics: the NPT’s Influence on Internal Decision Making 35 Robert Jervis, "Security Regimes," in International Regimes, Stephen D. Krasner, ed., (Ithaca:Cornell University Press, 1983), pp. 173-194. Indeed, the idea that the NPT is a successfulsecurity regime and thus an unexpected anomaly has been recognized by a few scholars. SeeRoger K. Smith, "Explaining the Non-proliferation Regime: Anomalies for ContemporaryInternational Relations Theory," pp. 253-281; Harald Muller, "The Internalization of Principles,Norms, and Rules by Governments: The Case of Security Regimes," in Regime Theory andInternational Relations, Volker Rittberger, ed., (Oxford: Clarendon Press, 1995); Allison L. C.


p. 36

This section outlines a political model of nuclear decision making and then uses this model to

explain how the NPT achieved the success it has enjoyed. There is no attempt to prove the

model—that will have to await another time—but there is an attempt to informally map out the

logic of the model and provide illustrative examples. To set the stage, the section begins with a

brief look at traditional approaches to thinking about non-proliferation.

A. Traditional Non-proliferation Analysis: Abstraction and Automaticity

Political science has traditionally been the academic discipline most concerned with nuclear

decision-making and non-proliferation. Ironically, the chief characteristic of non-proliferation

analysis, both past and present, is the absence of politics.36 Indeed, most writing on the subject

has been profoundly apolitical. Instead, nuclear decision-making has been treated as essentially

a mathematical process, where the decision to pursue or forgo nuclear weapons is a function of

security threats and technical capability.37

At its heart this approach embraces abstraction and automaticity. Given a certain level of threat

and sufficient technical prowess, the analyst posits that a country will “go nuclear.” Going

nuclear is an outcome without a process. It is as if it happens by fiat, and the actors themselves

play little if any role.38

de Cerreo, "Nuclear Non-Proliferation: the Origins of a Security Regime," Paper presented at the1994 Annual Meeting of the American Political Science Association, September 1-4, 1994.36 Some scholars have considered political variable. They point to the influence of competitionfor electoral advantage, what Solingen calls "coalition politics" and what Reiss refers to as"domestic pressures." On Sweden, Japan, and the role of domestic politics more generally, seeReiss, Without the Bomb: The Politics of Nuclear Non-proliferation, pp. 117-119, 249-251;Sagan, "Why Do States Build Nuclear Weapons? Three Models in Search of a Bomb," pp. 63-73.37 A recent US government commission on intelligence and weapons of mass destructionemphasized that one of the problems with American intelligence has been the failure toadequately account for political factors. The Commission on the Intelligence Capabilities of theUnited States Regarding Weapons of Mass Destruction, Report to the President of the UnitedStates, (Washington: GPO, March 31, 2005), pp. 13-14, 173-175.38 For a review of writing on threats and proliferation decision making, see Stephen M. Meyer,The Dynamics of Nuclear Non-proliferation; William C. Potter, Nuclear Power and Non-proliferation, (Cambridge, MA: Oelgeschlager, Gunn & Hain, 1982), pp. 131-144; Peter Lavoy,"Nuclear Myths and the Causes of Nuclear Proliferation," in Zachary S. Davis and Benjamin


p. 37

If the first irony is that this is political science without any attention to policy or politics, the

second irony is that most analysts, when describing their own country’s procurement or weapons

decisions, seldom fail to consider policy processes and political factors. Is the American desire

for missile defence purely a function of threat and capability? Many American analysts would

say no. They might cite bureaucratic politics, the unique role of Ronald Reagan (and Edward

Teller), political dynamics within the Republican Party, the timing and role of the Rumsfeld

Commission, the 1994 Republican takeover of the American Congress, or Bill Clinton and Al

Gore’s desire not to feed a perception that Democrats are weak on defence. Some of these

factors might be relevant, others not, but any analyst would be considered naïve if they looked

only at threats and capabilities and ignored the role of organizations, the interests of individuals,

and the workings of policy process. If those same analysts consider some other country’s

nuclear decision making, however, politics and policy are no longer considered relevant.

To be clear, the use of the word “politics” in this context is not restricted to, and indeed may not

even include, electoral politics, i.e., the politics of winning votes or support from the electorate

or public. Since most nuclear decisions take place in secret, most politics is not of the electoral

variety. It is more likely to involve the politics of organizations, the politics between countries,

and the politics among individuals, such as a prime minister or the head of the nuclear

bureaucracy.39

Frankel, eds., The Proliferation Puzzle: Why Nuclear Weapons Spread (and What Results),(London: Frank Cass, 1993), pp. 193-199; Scott Sagan, "Why Do States Build NuclearWeapons? Three Models in Search of a Bomb," pp. 54-86. In Meyer's survey of the literature,threat-related motivations were cited by eleven of twelve authors and constituted the most oftencited reason for wanting the bomb. Perhaps the broadest claim comes from Thayer, who assertsthat "security is the only necessary and sufficient cause of nuclear proliferation." Bradley A.Thayer, “The Causes of Nuclear Proliferation and the Non-proliferation Regime,” p. 486.39 See, for example, Lawrence Scheinman, Atomic Energy Policy in France Under the FourthRepublic, (Princeton: Princeton University Press, 1965). For more recent assessments, seeSagan, "Why Do States Build Nuclear Weapons? Three Models in Search of a Bomb." Thehistorical sociology of technology offers a similar perspective, often blending organizational andconstructivist elements. On historical sociology of technology and nuclear decision making seeSteven Flank, “Exploding the Black Box: The Historical Sociology of Nuclear Proliferation,”Security Studies, 3:2 (Winter 1993/94)); Tanya Ogilvie-White, "Is There A Theory of Nuclear


p. 38

To point out that politics and policy processes, like non-proliferation success, have been

consistently ignored is not to suggest that security and technical capacity are irrelevant. Both

factors can be decisive. In several cases, security has been a major, if not the major, proliferation

driver. This was likely true for Russia, China, Pakistan, and the initiation—if not the

continuation—of the United States nuclear weapons program in the 1940s.

Security and capability variables should always be considered, but as explanations for nuclear

decision-making, their reputation is greater than their explanatory power. Indeed, the problem is

that a singular focus on these aspects of nuclear behaviour has crowded out other, often more

promising, explanations. A myopic focus on security and capability has also led analysts to

ignore cases in which countries with few threats and modest capabilities nevertheless pursued

nuclear weapons or instances in which countries with ample threats and technical wherewithal

abstained from nuclear weapons. To truly understand nuclear behaviour, including the surprising

level of nuclear restraint, one must look to explanations unrelated to security: politics, pride, and

the role of domestic and international institutions.

B. A Political Model of Nuclear Decision Making

A political model of nuclear decision-making could take many forms, but the one offered here

begins with the fundamental premise that decisions favouring the acquisition or renunciation of

nuclear weapons affect the interests and preferences of organizations and individuals. Decisions

in either direction will benefit some (materially, ideologically, politically) and be costly to

others.

A second, related premise—one which finds widespread empirical support—is that every

country has a pro- and an anti-nuclear weapons constituency. This is not to suggest that the Proliferation? An Analysis of the Contemporary Debate," The Non-proliferation Review, 4:1(Fall 1996), pp. 43-60; and more generally, in the work of Donald MacKenzie and ThomasHughes, as in Donald A. MacKenzie, Inventing Accuracy: a Historical Sociology of NuclearMissile Guidance, (Cambridge: MIT Press, 1990); and Thomas P. Hughes, "The Evolution of


p. 39

factions in any given country are of the same size or equally powerful, only that there are

advocates on both sides of the issue. In some cases, these constituencies are well developed; in

other cases they are “latent.” In any case, the country’s nuclear future depends on which

coalition of constituencies is able to best use skill and chance to advance its agenda.

A political model does not imply that security threats or capability or alliances are irrelevant or

unimportant. Instead, it suggests that these factors are mediated by politics and the policy

process—that self-interested individuals and organizations understand, react to, and attempt to

use context and events to their own advantage based on pre-existing attitudes about the

desirability or legitimacy of these weapons.

In the political struggle between pro- and anti-nuclear constituencies, policy entrepreneurs on

both sides will seek allies. Both will seek to expand or narrow the scope of the political battle to

its benefit. In particular, they will seek to limit the composition of decision group in ways that

favour their chances of victory.40 Both will attempt to use external events or other “agenda

setting moments” and “policy windows” to advance their policy agenda. For example, pro-

nuclear advocates might use an unexpected nuclear test in another part of the world or

provocative military behaviour by a regional rival to argue that the question of nuclear weapons

should be “revisited.” Anti-nuclear advocates might similarly use the opening of a new arms

control treaty and its support by an allied superpower to push the government to reaffirm its non-

nuclear status. A change in government or in the leadership of a key ministry can also present an

opening for advocates to press their agenda.

Large Technological Systems," in The Social Construction of Technological Systems, Thomas P.Hughes, Wiebe E. Bijker, and Trevor J. Pinch, eds., (Cambridge: MIT Press, 1987).40 The decision group is the set of actors that actually decides policy. Sometimes it is acommittee (e.g., a primie minister and a defense minister or the cabinet); sometimes it consists ofa single person. On the importance of influencing the composition of the decision group, seeSchattschneider, who emphasized what he called the "scope" of political conflict. See E. E.Schattschneider, The Semi-Sovereign People: A Realists View of Democracy in America, (NewYork: Holt, Reinhart, and Winston, 1960).


p. 40

Pro-nuclear advocates are advantaged by secrecy, a monopoly over technical information, and

the seductiveness of nuclear technology. Opponents benefit from the institutional and economic

demands of a nuclear weapons program, which are substantially more costly than a chemical or

biological weapons program. As a percent of gross domestic product (GDP) or even of a mid-

sized defence budget, nuclear-weapon programs are not particularly costly, but as with any new

program, they are costly at the margin, i.e., they require funds in addition to all the other

budgetary commitments made by a government.

A key strategy for pro-nuclear advocates is to “keep the option open.” This allows them to bide

their time until that propitious moment when the inclinations of the head of state, the desires of

the relevant ministries, and external events all converge in a way that produce a favourable

policy opportunity. The advocates know that once the bomb is built, it is very difficult—though

not impossible—to reverse course.

Conversely, opponents push for a terminal commitment such as joining the NPT, i.e., a decision

to abandon nuclear weapons once and for all. They know that once such a commitment is made,

it is very difficult—though not impossible—to revive a bomb program.

In short, nuclear decision-making is first and foremost a struggle between competing players and

points of view. A nuclear weapons program does not happen magically on its own. Some

organization or individual must affirmatively push for their policy preference. There must be a

leader—a president, the head of the science or nuclear bureaucracy, an air marshal—someone

who “owns” the issue and is willing to spend resources, political capital, and attention to see it

through. Who are the likely leaders for such a cause? They are organizations that perceive that

they have the most to gain from a weapons program or individuals who have a particular

ideological or personal commitment to the idea of nuclear weapons.

According to this model, nuclear weapons decision-making is not unlike other policy decisions.

Indeed, one could look at nuclear weapons as a kind of procurement decision—which of course

it is. Military procurement has traditionally been viewed as a process that is drenched in politics


p. 41

of all kinds. Of course, acquiring nuclear weapons represents a special case of procurement. It

involves the highest level of government officials; it is secret rather than public; it involves the

most destructive weapon in human history. Still, the similarities with other examples of

procurement may make this analogy more helpful than misleading.

Viewed from the vantage point of a political model, the acquisition of nuclear weapons is

primarily an issue of political priority and sustainability. Is there a faction or individual who

cares enough about nuclear weapons that they will push the issue to the top of the policy agenda

and help keep it there for the ten-plus years it will take to complete the project? When it comes

to nuclear weapons, the engineering is more difficult than the science, program management is

more difficult than the engineering, and the politics is more difficult than the program

management. States that are successful in acquiring nuclear weapons do so, because they focus

on it, because they make it a priority, and because they finish the job before external events like

wars or changes in leadership derail the project.

Like any model, a political model of nuclear decision-making has its strengths, its weaknesses,

and its limits. A prudent analyst or policy maker should use different models to analyze the

same problem, taking what is useful from each perspective. One advantage of the political

model is that is helps explain some of the more curious aspects of proliferation. For example, it

can account for the fact that security threats are sometimes associated with pro-nuclear decision

making but that nuclear renunciation does not require the absence or even a reduction in threat.

Changes in political dynamics can also help explain variation in nuclear programs, in particular,

the way in which bomb efforts seem to wax and wane despite the fact that threats and capability

remain constant.

Perhaps most important of all, a political model helps explain the nuclear puzzle, i.e., the

unexpected success in restraining proliferation despite a less than perfect treaty, unfavourable

technological trends, and any number of crises. Exactly how a political model does this is

described in the next section.


p. 42

C. The Impact of the NPT on Internal Nuclear Decision Making

The power of the NPT is not that it addresses an individual country’s security threats. It does

not. The NPT’s constraining effect does not depend on a universal respect for international law

or adherence to a global norm. Instead, the NPT’s success in restraining the spread of nuclear

weapons is rooted in its power to reshape the internal politics of nuclear decision-making.

More specifically, the NPT constrained and influenced nuclear outcomes in at least seven

different ways.

C1. The NPT required that governments make a public declaration of their nuclear intentions.

Unlike most houseplants, nuclear programs grow best in the dark. Unfortunately for bomb

advocates, the NPT undermined one of their most important bureaucratic weapons: secrecy.

Secrecy can be used by pro-bomb factions to influence those who participate in the policy

process or to create a fait accompli that opponents find difficult to reverse.

Prior to the NPT, states were not obliged to reveal their nuclear intentions. With the NPT,

countries had to publicly declare their posture. Of course, a country was not required to sign the

NPT, but failure to do so brought unwanted attention, questions, and even suspicion.

Within the internal workings of governments, the NPT made a country’s nuclear weapons status

an open issue. Organizations and individuals that had a stake in the nuclear decision but had

previously been shut out of the decision process by secrecy were now aware that the question

was in play. The same was true for the public and for foreign capitals.

Secrecy was further undermined by the fact that once a country joined the treaty, it was obliged

to make its nuclear activities more transparent, e.g., by submitting to inspection.

C2. The NPT reframed the nuclear issue.

In the decades prior to the NPT, the nuclear issue was a defence issue. As such, the major

players were typically limited to the head of state and the military leadership (in some cases

prodded along by an ambitious nuclear or science bureaucracy).

With the advent of the treaty, a decision about nuclear weapons was no longer simply a defence

issue. It was now a foreign policy issue. Of course, issues of national security continued to play

a prominent role in the deliberations, but now other factors had to be weighed as well: the

attitude of superpower allies, regional relations, access to technology (Article IV), domestic

public opinion, international public opinion and status, and so on. The nuclear issue was no

longer “should we keep a nuclear option open?” but “should we sign a treaty to renounce nuclear

weapons that other countries are signing?” The treaty had the effect of conceptually and

politically transforming the nuclear question.

C3. The NPT influenced who sat at the table.

By reframing the nuclear decision, the NPT not only changed how the issue was evaluated but

also who could participate in the decision, i.e., it shaped the composition of the decision group.

It encouraged a decision group that extended beyond the prime minister, the military, and nuclear

weapons scientists to include the foreign ministry, treasury officials, and others. In most cases,

the effect of a) expanding the decision group and b) conceptually reframing the issue made it

more difficult for pro-bomb advocates to prevail.

C4. NPT ratification created bureaucratic winners and losers.

Government, by its very nature, prefers the status quo. Policy actors wishing to change policy or

pursue new policies usually have to exert more pressure to achieve change than those who prefer

that an existing policy continue. Of course, change is not impossible; governments do alter

policy, but change tends to be the exception rather than the rule—in part because the status quo

enjoys a presumptive status.

Policies reinforced by treaty commitments can make it even more difficult for policy

entrepreneurs, because the bar of presumption is raised to an even higher level. Again, reversal


p. 44

is not impossible, but it requires that proponents of a different policy spend more political

resources and witness even more extreme circumstances than would otherwise be required to

help their cause.

This general phenomenon applies to nuclear decision making as well. A little cited but path-

breaking paper by Rathjens, Chayes, and Ruina examines how the bilateral arms control process

between the USSR and the United States affected the internal politics of American nuclear

policy.41 A similar dynamic may be seen with respect to the NPT and a country’s internal

policy process.

Countries that wrestled with what to do about the NPT—and there were many—confronted two

options: join the treaty and renounce nuclear weapons or stay outside the treaty and retain a

bomb option. Keeping a bomb option did not mean ipso facto that a country would acquire

nuclear weapons, but joining the treaty entailed a commitment to renounce nuclear weapons.

Politically, these were asymmetrical choices, i.e., the decision to join the treaty was a more final

decision. It left little ambiguity. It created clear bureaucratic winners and losers.

A state could invoke Article X and withdraw from the treaty, but that was a legal matter. As a

political matter, having the issue decided and decided in a way that left little room for

interpretation meant that pro-bomb forces suffered a major if not mortal defeat. Having lost,

their chances of future victory were not zero, but they were not very high, and in any case, they

were certainly lower than they were before NPT ratification. Acquiring nuclear weapons was a

politically challenging and costly endeavour even without the treaty, but with the treaty, the hill

that bomb-advocates had to climb became higher and steeper.

C5. The NPT took away the “keep the option open” rationale.

The all or nothing character of nuclear renunciation helped create clear political winners and

losers, but this formulation had another effect as well. It took away one of the bomb advocates’

41 George W. Rathjens, Abram Chayes, and J. P. Ruina. Nuclear Arms Control Agreements:Process and Impact. (Washington, D. C.: Carnegie Endowment for International Peace, 1974.)


p. 45

most useful survival tools. Pro-nuclear policy entrepreneurs who had found themselves in a

hostile political environment were able to persist and fight another day by making the case for

“keeping the option open.”

Hedging arguments are difficult for policy makers to resist, because they give the (false)

appearance of costing nothing and providing flexibility for dealing with an uncertain future.

From a political standpoint, however, a “keep the option open” policy was simply a way for pro-

bomb actors to bide their time until a new election, a change in government, or external

circumstances gave them a better opportunity to pursue their agenda. Not only did the NPT

make bomb advocates political losers and thus put them in a disadvantageous position, it all but

foreclosed the hope that they could prevail at a later time when political circumstances were

more propitious.

C6. The NPT altered the character of post-ratification bureaucratic coalitions.

As indicated above, the NPT altered the political configuration of the decision group by framing

the nuclear issue in a way that brought additional, mostly anti-nuclear, constituencies to the

table. In other words, it helped reshape the pre-ratification political scorecard. Separately and

perhaps equally important, the act of joining NPT shaped the post-ratification political landscape

in a way that weakened pro-bomb advocates.

This effect was especially powerful in cases where militaries and nuclear bureaucracies had

colluded out of mutual self-interest to support a nuclear program. Prior to ratification, a nuclear

bureaucracy could dodge questions about the economics of civilian nuclear projects by a) using

the military dimension of the program to invoke secrecy rules that advantaged their position and

b) to push aside questions about cost by pointing out that their efforts had a national security

dimension.

After ratification, the military might withdraw from the political coalition, leaving the nuclear

bureaucracy to justify its ambitions on purely economic and civilian grounds—a task that was


p. 46

often beyond their capability. This, in turn, would weaken the nuclear bureaucracy’s political

and programmatic position.

In the case of Australia, for example, the common nuclear agenda shared by the Ministry of

Defence and the Australian Atomic Energy Commission (AAEC) fell apart after NPT

ratification. Prior to ratification, the Ministry of Defence and the AAEC (and most especially its

chair, Philip Baxter) had supported the common goal of a nuclear weapons capability and

provided political backing for each other’s programs and positions in various inter-agency

deliberations. Following ratification, with weapons acquisition no longer a live possibility, the

military saw no reason to support the AAEC’s programs. As a result, the AAEC’s projects were

evaluated solely on the basis of economic viability. The combination of fewer political allies and

heightened scrutiny proved too much. Outside of mining and materials, the Australian nuclear

program was essentially halted. In short, the treaty helped break up the pro-nuclear alliance and,

in doing so, severely weakened one of the advocates of nuclear acquisition. Both effects

increased the likelihood of treaty compliance.

C7. NPT treaty commitments were progressive and sticky.

One of the NPT’s greatest (and unrecognized) strengths is that it is dynamic. Unchanging

general principles are encoded in the treaty itself, but the application of those principles is found

in safeguards agreements, which have evolved over time. That evolution has been

unidirectional: safeguards have only gotten stronger with each iteration. This process creates a

situation in which states become inextricably entangled in progressively stronger constraints on

their nuclear programs.

A given government might refuse to sign the treaty, but the treaty does not disappear. It

patiently lies in wait. Eventually a government comes to power and decides to join the treaty,

and then the country is committed. Moreover, the safeguards-arrangements that a government

agreed to in 1975 are not the safeguards arrangements they have to abide by in 2005. An initial

commitment to join the treaty makes it politically easier for another government to commit to


p. 47

stronger safeguards. Having committed to those arrangements, it becomes easier to commit to

still stronger safeguards.

This is not to suggest that all states have welcomed new safeguards. Indeed, those burdens are

often resisted. From an aggregate or system perspective, however, the trend line is clear: serial

commitments to progressively stronger safeguards. This result is, in part, a political dynamic.

Commitments are settled policy. Acquiescence to new safeguards-agreements is consistent with

that settled policy; it does not represent a new policy. And with each new commitment to a

slightly stronger safeguards agreement, the bar is raised still higher for pro-bomb advocates.

In short, a single decision by one government can put a country on a track that, twenty years

down the line, delivers it to a very different place. Particularly vivid evidence of this process

could be seen at the 1995 NPT Review and Extension Conference. Many of the countries that

originally expressed reservations about the treaty—Germany, Australia, and Italy, among many

others—were now its strongest proponents. One only has to read the speeches from 1969 and

compare them with the speeches from 1995 to see how dramatically the centre of gravity had

shifted.42

VI. Lessons (Un)Learned: the NPT’s Past Success and Today’s Non-proliferation

Challenges

This inquiry began with the premise that policy makers should give special attention to past

success. The paper then documented the ways in which the NPT has proven to be an

unexpected but powerful constraint on the spread of nuclear weapons. In the previous section, 42 On Italy’s nuclear past, see Cesare Merlini, "A Concise History of Nuclear Italy," TheInternational Spectator, (Journal of the Istituto Affari Internazionali, Rome), 23:3 (July-September 1988); Leopoldo Nuti, "Italy and the Nuclear Choices of the Atlantic Alliance, 1955-63," in Securing Peace in Europe, 1945-62, Beatrice Heuser and Robert O'Neill, eds., (London:Macmillan Press, 1992); Steven J. Baker, Italy and the Nuclear Option, (Santa Monica:California Arms Control and Foreign Policy Seminar, 1974). On Germany, see CatherineKelleher, Germany and the Politics of Nuclear Weapons, (New York: Columbia University


p. 48

the reasons for that success were investigated and a new model, a political model, of non-

proliferation was outlined. In this section, the contemporary challenges that confront the non-

proliferation regime are inventoried and then re-evaluated given the lessons of past success.

A. Today’s Non-proliferation Challenges

Despite the periodic lament that the non-proliferation regime is dying or dead, the NPT continues

to constrain the spread of nuclear weapons. Nevertheless, the regime faces significant tests,

some immediate and some further down the road. The list of dangers offered here is presented in

no particular order.

1. Fuel cycle technology in general, Iran in particular

2. Nuclear terrorism and the problem of unsecured bomb material

3. Nuclear black markets and private networks

4. Resistance of the nuclear weapons states

5. Belief that the “NPT is no longer relevant”

6. Nationalism (e.g., United States, Iran)

7. IAEA’s obligations versus its capacity

Students of non-proliferation are certainly familiar with some of the challenges on this list. Fuel

cycle issues, that is, the problem of unconstrained access to reprocessing and enrichment

technology, have been widely discussed. Similarly, the dangers posed by nuclear terrorism and

nuclear black markets have been placed on most policy agendas as a result of the 9/11 attack and

the activities of the Khan network.

Other items on the list are probably self evident even if they are not as widely discussed. The

belief that the regime is no longer relevant undermines the regime, both because it may

encourage some states to reconsider their nuclear options, and because it encourages other states

Press, 1975), and Jefferey Boutwell, The German Nuclear Dilemma, (Ithaca: Cornell UniversityPress, 1990).


p. 49

to take extreme actions that may serve to undermine the regime, e.g., preventive war, the

development of new nuclear weapons.

The problem of the institutional capacity of the IAEA is not particularly sexy, but long-time non-

proliferation officials and scholars know that the responsibilities of the agency have grown

exponentially, while in general, its resources have remained the same.43 So far, the effect of this

gap has not made itself felt in a particularly dramatic way. Still, the situation is undoubtedly

unsustainable in the long term. If the resource-to-responsibility imbalance is not addressed, then

the agency will likely fail at some point, and its effectiveness will be called into question.

Other aspects of this list may be surprising. It appears not to mention North Korea or Israel. It

does not include the problem most often cited by many American analysts: non-compliance and

the need for enforcement.44 And what is this business about nationalism?

Actually, North Korea and Israel are included. They are subsumed under number 4 along with

India, Pakistan, and the recalcitrant “official” nuclear weapons states. Though they represent

different kinds of problems,45 what all these states have in common is a reluctance to give up

their nuclear status, or as in the case of the United States and Russia, a desire to augment their

nuclear status or the role of nuclear weapons in their defence doctrines. These countries’ nuclear 43 See for example, MIT Working Group on Nuclear Materials Management, "Options forPreventing Illicit Trafficking in Fissile Materials," (Cambridge: MIT Security Studies Program),1995; Kaleb J. Redden, “Inspecting the Inspectorate: A Look at Financial and Political Supportfor the IAEA,” Non-proliferation Review, 10:3 (Fall-Winter 2003), pp. 34-47.44 The American focus on compliance is reflected in the statments of both officials andnongovernmental non-proliferation analysts. See, for example, Stephen G. Rademaker,Testimony, Hearing on the Nuclear Non-proliferation Treaty, US House of Representatives,International Terrorism and Non-proliferation Subcommittee, International Relations Committee,April 28, 2005, Congressional Quarterly; Jean Du Preez, Testimony, Hearing on the NuclearNon-proliferation Treaty, US House of Representatives, International Terrorism and Non-proliferation Subcommittee, International Relations Committee, April 28, 2005 CongressionalQuarterly; Michael A. Levi and Michael E. O’Hanlon, The Future of Arms Control,(Washington: Brookings Institution, 2005).45 With the DPRK, the issue is compliance with its safeguards obligations; with the “offiical”nuclear weapons sttes, the issue is complance with Article VI; and with India, Pakistan, andIsrael, the issue is their status as nuclear weapons states outside the treaty.


p. 50

attachments have different effects. North Korea’s nuclear weapons put proliferation pressures on

other countries in the region. The pro-nuclear positions of the United States, and to a lesser

extent Russia and China, make political agreement on ways to enhance the regime difficult if not

impossible, e.g., lack of action on the so-called “13 Steps” enumerated in the 2000 NPT Review

Conference.

Non-compliance and enforcement are not included, because they are not the main problem. In

general, compliance and the current system of discretionary bilateral and Security Council

enforcement have worked very well. Moreover, as suggested above, compliance has more to do

with internal politics than mechanistic threats and penalties.

Nationalism is included, because there appears to be a very strong connection between

nationalism and nuclear weapons. Nationalistic leaders—de Gaulle, Ben Gurion, Gorton,

Nasser, Sukarno, Zia to name a few—seem to be the kind of leaders most interested in nuclear

weapons. In addition, countries undergoing a nationalist wave seem most susceptible to the

allure of nuclear weapons, whether it is India, the U.S. after 9/11, or Iran today. Nationalism

often entails a desire for power and prestige, a reluctance to be constrained by multilateral

agreements, and a “go it alone” defence philosophy—none of which are particularly conducive

to non-proliferation. Countries caught in the grip of nationalism provide pro-nuclear advocates

with a friendly environment in which to wage their political battles.

The response to these challenges—at least the ones that command public attention, such as the

fuel cycle, terrorism, and black markets—has been to suggest various policy fixes, e.g., ban

reprocessing, secure nuclear material, or arrest the marketeers. With rare exceptions, there has

been little focus on how these remedies might be adopted. In other words, there has been

virtually no discussion of the politics of these proposals and what is required if a diverse set of

countries is going to cooperate in their implementation. The next sections provide a preliminary

look at these issues. Before turning to the individual threats to the regime, however, attention is

given to the lessons of the NPT for non-proliferation policy in general.


p. 51

B. Lessons from the NPT Experience for Non-proliferation Policy in General

The success of the NPT and the non-proliferation regime offer a number of different lessons for

policy makers. Before turning to the lessons, however, it is important to be clear about what is

not a lesson.

B1. Treaties are not always the answer.

Given the analysis so far, it might be tempting to conclude that treaties are the answer to the

problem of non-proliferation. Such a conclusion would be wrong, however. The NPT was one

of the most important, if not the most important, development in restraining the spread of nuclear

weapons. That does not mean that going forward, a treaty is always the answer. Under some

circumstances it may be the appropriate policy remedy; on other occasions, it may not be.

Treaties are comparatively blunt and rigid instruments. In rare instances, they can even spur

proliferation behaviour.46 They are best suited to the circumstances like those of the NPT, i.e.,

trying to influence the behaviour of a large number of countries all at one time and prohibiting or

denying (rather than regulating) a technology. As it stands, virtually every country is already in

the NPT, so the special utility of treaties may be somewhat limited.47

Instead of focusing on treaties per se, it may be more useful to give attention to the underlying

reasons why this treaty, the NPT, proved so effective. It is these more general lessons that may

prove the most help to policy makers.

46 This can happen, for example, because a treaty creates an agenda setting moment (i.e., nuclearpolicy is suddenly open for action) at a time when the decision group in power is tilted in favorof weapons acquisition. Australia’s interest in nuclear weapons acquisition in response to theTest Ban talks provides one example. It has also been suggested that the CTBT contributed toIndia’s decision to test in 1998. On Australia and the Test Ban, see Australian Archives (ACT):A5818/2, Robert Menzies to the Cabinet, Nuclear Tests Conference: Control Posts in Australia,Submission No. 1156, V6, p. 13 (Secret); Australian Archives (ACT): A1838/269,TS852/10/4/2/3; Letter from Prime Minister Menzies to Prime Minister Macmillan, 29th June,1961, p. 1 (Secret). On India and the CTBT, see Dinshaw Mistry, "Domestic-InternationalLinkages: India and the Comprehensive Test Ban Treaty," Non-proliferation Review, (Fall1998), pp. 25-38. More generally, see Fred Charles Ikle, “Nth Countries and Disarmament,”Bulletin of the Atomic Scientists, December, 1960.47 Nevertheless, the CTBT or the FMCT would certainly be welcome additions to the non-proliferation regime.


p. 52

B2. Attention to internal politics.

The first and most important lesson is that those who wish to prevent the spread of nuclear

weapons must be attentive to the internal political and policy dynamics within states.

The overwhelming tendency in non-proliferation policy making, particularly as regards

American policy, is to assume that U.S. decision-making is the most important variable in

proliferation outcomes. In the United States, both liberals and conservatives suffer from Beltway

myopia. The natural consequence is to place less focus and therefore to know little and

understand less about the political context in other countries, where the actual decisions are

being made.

This first tendency helps foster a second that has the same effect. Non-proliferation supporters in

the United States—officials, nongovernmental organizations, foundations, academics—attempt

to craft non-proliferation remedies in the abstract without regard to the political consequences

they will have in the countries they are trying to influence. This technical approach, which is not

immune from national bias, often results in proposals that have no hope of adoption or that once

adopted can even have counter-productive effects.

The irony is that non-proliferation advocates in the United States are extremely sensitive to the

politics of non-proliferation policy in their own country (e.g., the president’s popularity rating,

how a legislator from Ohio has voted in the past, the state of interagency warfare concerning a

particular proposal), and yet they know virtually nothing about the countries at whom their

policy remedies are directed. If they took the same attitude towards U.S. policy making that they

do to other countries, their colleagues would accuse them of being naïve, irrelevant, or even

dangerous.

B3. Support non-proliferation constituencies.

If the first lesson is that policy makers need to understand the political and policy processes of

nuclear decision making in countries they care about, then the second lesson is that they should


p. 53

use this knowledge to develop and strengthen the position of non-proliferation constituencies

and/or prevent or undermine the development of pro-nuclear constituencies. This is the

fundamental principle that should guide non-proliferation policy. The distribution of powerful,

informed, and active domestic non-proliferation constituencies—treasury officials worried about

boondoggles, nuclear scientists who oppose military interference, army chiefs of staff sceptical

of how nuclear weapons actually help the army rather than the air force, investigative journalists

looking for a story, an anti-nuclear public, a prime minister or presidential candidate who has a

long record of opposing proliferation—is the surest guarantee of preventing proliferation or,

failing that, delaying or discovering the actions of a pro-nuclear faction.

This principle of developing and strengthening the ability of non-proliferation constituencies

(and doing harm to their opponents) actually entails two different processes. Building or

developing non-proliferation constituencies is one task. Strengthening the political position of

existing non-proliferation constituencies so that they can prevail in their internal policy battles is

a second and separate task. Both are important, but the focus here will be on the latter as it is

directly related to the lessons of the NPT experience.

As noted earlier, non-proliferation constituencies can be helped or hindered by the way in which

a nuclear issue is framed, e.g., weapons acquisition as a defence issue versus a foreign policy

issue. Framing can define what factors are considered relevant to the decision and how they are

weighted. In addition, framing has the absolutely critical ability to determine what parties get to

participate in the decision. The objective should be to frame an issue in such a way that it

maximizes the participation of anti-nuclear weapons constituencies in a manner that plays to

their strength while at the same time minimizing the participation and political position of bomb

advocates.

This is not a matter, therefore, of good public relations (which is nevertheless always welcome).

It is not about deciding on some remedy or policy fix ex nihilo and then thinking like a political

consultant to sell the idea. Instead, this is about designing or structuring the policy

instrumentality in a way that it allocates benefits and costs so as to draw the attention of the


p. 54

target constituencies and motivate their participation. After designing an instrumentality based

on a political model, the task is to take that remedy and frame it in a way that maximizes its odds

of success.

Article IV of the NPT, for example, helped recast the interests of some civilian nuclear

bureaucracies from pro-bomb to pro-non-proliferation. Prior to the NPT, a struggling civilian

nuclear bureaucracy might find the prospect of a weapons program a welcome solution to its

budgetary woes. Article IV changed that when it specified that parties to the treaty would enjoy

the fruits of nuclear technology. Some national nuclear organizations inferred that failure to join

the NPT would result in a denial of access to nuclear technology. Fearing that the denial of

nuclear technical assistance would doom their programs, some organizations switched

allegiances and became treaty (i.e., non-proliferation) advocates.

This example is noteworthy both for the political dynamic it illustrates and because of the

organization it highlights, namely, civilian nuclear organizations. Nuclear bureaucracies, along

with armies and treasuries can be particularly important non-proliferation policy players.

Civilian nuclear organizations can be critical “swing votes” in nuclear decision-making. They

can be avid supporters of bomb projects for reasons of pride or bureaucratic self-interest. They

can also side in favour of non-proliferation for reasons of morality or bureaucratic self-interest

(e.g., fear of military interference in their scientific work). It is almost impossible to have a

successful bomb program without a committed nuclear organization. (The historical record on

military-only initiatives is not one of success.) Moreover, the nuclear bureaucracy has an

information advantage over other bureaucratic players in terms of estimating whether a particular

technological goal is achievable and at what cost.

Two other organizations merit mention, if only because they are so often overlooked.48 One is

the treasury. These keepers of the purse are genetically predisposed to oppose new spending 48 More generally, potential non-proliferation constituencies may include the head of state, theforeign ministry, the army, the cabinet, the legislature, the trade or commerce ministry, thetreasury, the civilian nuclear bureaucracy, the energy ministry, the scientific establishment, thenuclear industry, the business community, the mining industry, the private energy sector, the


p. 55

projects, especially on something as exotic and risky as nuclear weapons. Historically, they have

not been as powerful as they might be, however, because bomb advocates used secrecy to bypass

their reviews, and because they lacked the technical information to properly assess the claims of

pro-nuclear coalitions. Nevertheless, they are a natural non-proliferation constituency, and so it

makes sense to a) create conditions in which they are likely to participate, and b) insure that they

have the training and technical information to cut through the fog of the pro-bomb group’s

promises.

Another potential non-proliferation constituency is the army, as distinct from the air force or

navy. Indeed, inter-service rivalry can be a powerful impediment to proliferation.

Unfortunately, it can be a powerful spur as well. For the soldier in the field, a nuclear weapon

does nothing, except maybe kill him or her along with the enemy. From a soldier or general’s

standpoint, an extra tank or a better supply line has a more direct impact on the ability to prevail

in battle. Moreover, if the money for a nuclear weapons program is drawn from operations and

maintenance budgets, the soldier in the field actually does worse with a nuclear weapons

program. Contrast the army’s situation with that of the air force, which is likely to receive new

planes and related resources in order to deliver the nuclear weapons, or to a lesser extent, the

navy, which may hope to develop nuclear propulsion.

Of course, there are cases where any army has supported a nuclear weapons program.

Sometimes, a particular army chief is simply pro-nuclear. On other occasions, the army does not

clearly see its self-interest and is log rolled by the other services. More often, the army is

“bought off” with the promise of being the service that will be in charge of the nuclear program.

At its worst, inter-service rivalry can encourage proliferation, as when each service demands and

gets its own nuclear program.

health sector, environmentalists, journalists, universities, political parties, and the public. Thoseindividuals or groups most likely to support nuclear weapons development include the head ofstate; the military and in particular the air force, occasionally the navy, the ministry of supply orthe defense industrial organization, and the defense science board; the cabinet, individual scienceadvisors), the civilian nuclear bureaucracy, nationalist political parties, and a nationalist public.


p. 56

B4. There must be benefits in non-proliferation.

Implicit in the last two points is the notion that there must be benefits to non-proliferation.

Certainly, there are costs, and therefore, there will be domestic players who oppose joining or

extending the non-proliferation regime. These costs must be outweighed by the benefits, or more

precisely, the benefits of non-proliferation must be sufficient that they can catalyze the

participation of enough non-proliferation advocates for non-proliferation policy to win.

Benefits can take several forms: material, political, ideological, or social. Part of the NPT’s

success can be attributed to the fact that it provided or was perceived as likely to provide a

variety of benefits, including everything from civilian nuclear assistance to membership in the

club of responsible nations. Indeed, as is often said, the NPT was a bargain between nuclear

weapons states and non-nuclear weapons states.

Today, that bargain does not have the luster as it did in 1970. To begin with, the nuclear

weapons states have not followed through on their Article VI obligations, and the most powerful

of the nuclear weapons states has blocked progress on the “13 Steps” of the 2000 NPT Review

and Extension Conference. It also has to be said that civilian nuclear technology and in

particular, nuclear energy, did not fulfil the promise or predictions of the 1950s. Energy has not

become too cheap to meter. Most countries looking for cutting edge technology are likely to be

more interested in information technology or biotechnology than nuclear power. Recently, there

appears to be more interest in nuclear energy, but it is unlikely that it will ever yield the kind of

benefits its promoters envisioned decades ago. Accordingly, it is imperative that the bargain be

revitalized—that nations see material and other benefits that outweigh the material costs of

safeguards or the economic costs or not siding with a powerful country that has pro-nuclear

aspirations. A revitalized bargain may necessitate going beyond nuclear technology to other

kinds of benefits, e.g., energy assistance in general, scientific technical cooperation, or more

generic incentives.

The payoff for vigorously supporting non-proliferation should be high enough for at least a few

states to take the non-proliferation mission as a national identity. Japan, the Nordic countries,


p. 57

and Australia have, at times, taken on this role. Put another way, non-proliferation cannot solely

be about restraining “bad” countries. If the regime is to strengthen over time, it needs its own

advocates, countries willing to expend political energy for non-proliferation goals.

Finally, a non-nuclear status should be seen as a viable political strategy that has potential

national policy payoffs for countries. Egypt, for example, rightly chose to foreswear nuclear

weapons and instead has affirmatively used its non-nuclear status as part of a political strategy to

constrain Israel’s nuclear weapons program. This is exactly what non-proliferation advocates

should want, but it is only sustainable if countries such as Egypt receive some satisfaction for

their efforts. In this particular case, that means giving greater attention to the problem of Israel’s

nuclear arsenal. In short, non-proliferation advocates should hope that more states see non-

nuclear status as an effective alternative for addressing their security and political needs.

B5. Transparency.

As previously observed, secrecy tends to favour pro-bomb advocates, whereas transparency can

advantage bomb opponents as well as provide the kind of early warning that will slow or

complicate the pro-bomb agenda. Again, it is important to emphasize that success requires not

only that non-proliferation constituencies are strengthened but also that pro-bomb entrepreneurs

are put at the maximal possible disadvantage. To date, the history of NPT is the history of

transparency qua safeguards agreements making life difficult for aspiring or potentially aspiring

nuclear weapons states (e.g., Iraq post-1991, North Korea in the early 1990s, and Iran after

2002). Of course, formal arrangements are but one part of transparency. Investigative

journalists trained in nuclear issues, environmental groups, legislatures, dissident groups, and

evolving technology can contribute to broader and deeper transparency in a state’s nuclear

affairs.

B6. Promote progressive commitments.

As discussed earlier, the continuing evolution and improvement in safeguards arrangements has

resulted in a system that makes it more difficult over time for states to rethink their non-nuclear

status. While few countries welcome taking on more obligations, history suggests that new


p. 58

iterations of safeguards obligations will be steadily, if gradually, accepted by NPT member

states, especially following a crisis or controversy over a problem state. Policy makers would be

wise to continue to develop future safeguards standards—behavioural as well as technical—and

thus further operationally and politically constrain pro-nuclear entrepreneurs. In short, leading

states and NGOs should already be working now on whatever will follow the Additional

Protocol in the future, even if member states are not yet ready to consider the new burdens it

would entail.

B7. Conduct political diagnoses and watch for agenda setting moments.

Lists of potential proliferators are a staple of the non-proliferation business. In every decade

since Hiroshima, analysts and academics have produced a scorecard of countries they believe are

likely to be tempted by the bomb. Some of the countries turn out to harbour nuclear ambitions,

some do not, and others not on the list have nevertheless pursued nuclear weapons. The basis for

these assessments is almost always some combination of a country’s technical capability (i.e.,

how advanced is their nuclear infrastructure) and an assessment of the security threats it faces. A

political model of nuclear decision-making would suggest that these efforts should be

supplemented by a different kind of exercise.

Rather than beginning with threats and capabilities, a political assessment would begin with a

profile of the nuclear preferences of the key individuals and organizations, an assessment of their

decision power or influence, and a map of the decision making process (e.g., who participates

and when). Ideally, the analysis would include an inventory of past nuclear decision-making

and/or a profile of potential future players in the policy process (e.g., following an election, a

retirement, or a coup).

This application of the political model would be especially sensitive to future “agenda setting

moments.” These agenda setting moments consist of occasions when nuclear policy becomes a

live or actionable issue for a government. Typically, governments do not consider major

changes in nuclear policy every day. Instead, proposals work their way up the policy process or

are event-driven. For example, a nuclear test or a new nuclear treaty or a change in government


p. 59

can be agenda setting moments—situations in which some set of players is called together to

consider or reconsider a nation’s nuclear policy. Focusing on the nuclear preferences of specific

policy actors, the nature of the policy process, and the moments at which nuclear policy becomes

subject to redefinition may provide a more powerful way to assess proliferation risks than a

simple tally of reactors and enemies.

C. Lessons Applied to Today’s Non-proliferation Challenges

This section began with a review of some of the more pressing challenges that confront today’s

non-proliferation regime. Having explored the lessons of the NPT’s success for non-

proliferation policy in general, the focus now turns to the application of those insights to those

specific non-proliferation threats. Given limited space, two of the seven will receive special

attention: fuel cycle issues and nationalism. These two, in particular, illustrate the potential

value of the political model on nuclear decision-making.

C1. The problem of the fuel cycle.

The question currently confronting policy makers is how to persuade states to forgo what, under

the treaty, they are entitled to, namely, access to reprocessing and enrichment for civilian

purposes. The lessons of the NPT seem particularly well suited to this problem. First, consider

the organizational politics of the fuel cycle. As it now stands, national nuclear organizations will

likely oppose the renunciation of their country’s fuel cycle rights, as this would represent a limit

on their programs and possibly a decline in their budgets. To date, the only benefits offered for

an alternative fuel cycle arrangement involve a guaranteed fuel supply, but this benefit accrues to

the country as a whole, not the domestic nuclear organization. The national nuclear

bureaucracy’s opposition to renouncing the fuel cycle will likely be supported by any elements in

the country’s military that still harbour nuclear ambitions or who hope to acquire a hedge in the

form of a complete fuel cycle. Add to that the fact that some of the countries most interested in a

fuel cycle are in the midst of a nationalist phase (e.g., Iran), and it should be no surprise that

hectoring by the United States and others about the dangers of the fuel cycle has not changed

many minds.


p. 60

Success on the issue of the fuel cycle will require that domestic nuclear bureaucracies perceive a

reason, a benefit, in giving up something they view as valuable. Success will also be more likely

if incentives are offered that broaden the scope of participation to include treasury officials and

foreign ministries.

More generally, the entire issue has to be reframed. The current formulation plays badly for non-

proliferation advocates insofar as the issue is seen in many quarters as one of a) protecting rights

granted under a treaty and b) yet another attempt by rich countries to deny technology to poorer

countries in a way that is discriminatory. These are two powerful emotional and political

rationales for the pro-fuel cycle faction. The anti-fuel cycle faction, by contrast, must rely on

what must seem like vague concerns about a proliferation danger in a distant future involving

countries that have already pledged to forgo nuclear weapons.

Supporters of non-proliferation would do well, given this context, to avoid the U.S. tactic of

public pronouncements of technology denial that likely come across as coercive and unilateral.

Such pronouncements only confirm the sceptics’ doubts about the true objectives of fuel cycle

reform.

A better strategy would be to reshape the debate—perhaps from proliferation to energy and

development—and come up with a new package that promises energy benefits to participating

countries and provides national nuclear organizations with a reason to participate. This energy

initiative would extend beyond nuclear technology alone, i.e., the nuclear component would be

subsumed into a broader concept. This would bring new organizational constituencies into the

decision process and would remove the “rich versus poor” frame in which the issue is now

entangled. The focus of the package would be energy development and assistance, but one of the

requirements would be a suspension of the right to reprocessing and enrichment. Presumably,

the number of takers would be small at first, but if they were well rewarded and there was a

deadline for joining, other states would likely follow course rather than be left out. If the deal

were to combine larger benefits up front with progressive commitments to non-proliferation over

time, countries might be drawn in and then find it hard to reverse course.


p. 61

C2. The problem of nationalism.

Nationalism poses an extraordinarily difficult and dangerous challenge for non-proliferation.

Most countries go through nationalist periods, and so it is especially important for non-

proliferation advocates to be clear about the most appropriate tactics for responding to

nationalism so that the risks of proliferation are kept to a minimum. Here again, the experience

with the NPT provides some useful guides.

Virulent nationalism provides bomb advocates with a permissive political environment. The

heady passion of nationalism can lead officials to favour policies for which the costs are higher

than the benefits, even when their own organizational or personal self-interest suffers as a result.

Under these conditions, a strategy based on eliciting the support of specific non-proliferation

constituencies is less likely to yield positive results. Indeed, expanding the scope of the internal

battle over nuclear policy may actually result in a stronger pro-bomb coalition.

Instead, a political model of nuclear decision-making would probably point to three other

alternate approaches. The first is to simply wait for the fever to pass. Do not press ahead with

non-proliferation strategies that force the issue, i.e., create agenda setting moments that result in

policy choices that become fixed policy. Doing so will likely produce outcomes that are the

opposite of what is being sought and put domestic non-proliferation advocates in a worse

position. The history of the NPT was not that all countries joined the treaty once it was open for

accession. Many countries waited years, even decades before joining. Once in the treaty

however, bomb advocates found themselves progressively constrained from pursuing their goals.

A second but more challenging tactic is to reframe the non-proliferation debate in such a way

that a commitment to non-proliferation is consistent with nationalism, e.g., that the country in

question receives status benefits or enhanced prestige as a consequence of their actions. South

Africa enjoyed something akin to this following the voluntary dismantlement of its weapons

program. It might be argued that Japan, Sweden, and Egypt have enjoyed some status benefits

for being leading voices in opposition to nuclear weapons. In this context, it is worth


p. 62

remembering that benefits need not be restricted to material goods such as aid or energy projects.

Social benefits can also be used as incentives. A leadership position in an international forum,

hosting a visit from the president of the United States, high profile individual appointments, or

even awards can confer the respect and prestige that nationalist policy makers seek.

Finally, international non-proliferation advocates would do well to avoid public condemnation of

countries undergoing a period of nationalism. Such criticism will only intensify feelings of

nationalism and undermine the ability of indigenous non-proliferation advocates to positively

influence domestic debates.

None of these options is easy or very satisfying, but sometimes not fighting is better than

fighting and losing. A political model of decision-making highlights the fact that timing does

matter. More generally, international non-proliferation advocates need to be able to recognize

when the moment is right to push ahead (e.g., in the wake of a crisis when the appetite for action

is high) and when to be patient. Like a good politician, they need to be able to count votes, to

force the issue when they are ahead, and to delay when they are not.

C3. Other threats to the regime.

The lessons of the NPT and a political model of nuclear decision making have implications for

the other threats facing the regime as well. As regards the security of nuclear materials and

nuclear terrorism, for example, the importance of benefits and finding organizational allies

would seem to be relevant. The NPT experience would also appear to suggest that it is

preferable to prohibit something rather than regulate it, and thus the desirability of proposals for

a “global cleanout” of special nuclear materials, a fissile material cut-off treaty, or the

establishment of “fissile material free zones.”

Another example is the resistance of nuclear weapons states. Consider the United States. A

political diagnosis of the problem would point to the need to decouple the national nuclear

laboratories’ and Department of Defense’s organizational interest in maintaining or even

expanding nuclear arsenals. Already, some voices in the Pentagon have begun to complain that


p. 63

nuclear weapons inhibit the ability of the American military to achieve other important

objectives, e.g., when budgets are tight and well equipped troops are more important that a

nuclear warhead. In each of the weapons states, the political landscape will be different, but

international non-proliferation advocates need to address the material, political, and social

incentives that foster resistance to non-proliferation and disarmament—that is, finding ways to

reshape the incentive structure and then preparing for and acting when a favourable agenda

setting moment presents itself.

D. Limits to the Lessons

The lessons of the NPT and the related insights of a political model of nuclear decision making

can substantially improve the prospects for future non-proliferation success, but nothing said

here should be interpreted as suggesting that these are magic bullets or that other non-

proliferation tools and perspectives can be tossed aside. The problem is sufficiently complex and

the individual circumstances sufficiently diverse that no one approach or tactic will be enough.

So for example, the political model, while still useful, is less robust when applied to the problem

of nuclear black markets than it is to the problem of the fuel cycle.

There are other limits as well. Many of the dynamics described here may work more effectively

vis-à-vis democracies than authoritarian states, though it must be emphasized that authoritarian

states have organizational factions and politics. Moreover, some of the countries on non-

proliferation watch lists—Taiwan, South Korea, Brazil, and Japan—are democracies. Still,

every model has its limits and will perform better in some environments than in others.

The key point, however, is that ignoring the success of the NPT and the political model of

nuclear decision-making is a serious error. Today, the danger is not that non-proliferation

specialists will so narrowly focus on the NPT and political factors that they will ignore security

and technical variables. No, the more likely mistake—one widely observed in the past and

present—is that they will wholly focus on the latter to the exclusion of the former.


p. 64

VII. Conclusion: The Three Questions

This inquiry began with three questions. Has the NPT been a success? Why has it been a

success? What does this success tell policy makers about how to restrain (and even reverse) the

spread of nuclear weapons, and in particular, how to approach the contemporary challenges that

face the non-proliferation regime.

The answers provided have been somewhat unconventional. Despite the doom and gloom

coming from many quarters, this analysis maintains that, by most any standard, the NPT has

been an unexpected and unheralded success. In addition, it has been suggested that much of this

success can be traced to the impact of the NPT on the internal policy processes of states, and that

if that success is to be preserved and even expanded, policy makers must focus on the

development and strengthening of indigenous non-proliferation constituencies, attention to

agenda setting moments, as well as the use of transparency, incentives, and an ever evolving

application of non-proliferation instrumentalities.

It is also worth noting what this paper does not say. It does not suggest that because the NPT

was able to work non-proliferation magic, new treaties are necessarily the most important tools

in the non-proliferation toolbox. It does not declare that just because the NPT has been a success

everything is rosy. Indeed, it identifies a number of threats to the regime, some commonly cited,

others less so. Finally, the analysis does not contend that security and technical capability are

unimportant and never decisive.

What is suggested, however, is that a myopic preoccupation with security and technical capacity

has diverted attention away from other equally important considerations, particularly the role of

politics and the importance of nationalism. More generally, this paper has attempted to offer a

different way of thinking about the problem of proliferation and alternative approaches to

restrain and reverse it.

Finally this paper ends where it began—with the question of success. It has insisted that the

phenomenon of success is important, that it has been systematically overlooked and undervalued,


p. 65

and that if policy makers and scholars continue to ignore the reality in front of them, they will

put the regime and global security at even greater peril.

No 1 “Review of Recent Literature on WMD Arms Control, Disarmament and Non-Proliferation” by Stockholm International Peace Research Institute, May 2004

No 2 “Improvised Nuclear Devices and Nuclear Terrorism” by Charles D. Ferguson and William C. Potter, June 2004

No 3 “The Nuclear Landscape in 2004: Past Present and Future” by John Simpson, June 2004

No 4 “Reviving the Non-Proliferation Regime” by Jonathan Dean, June 2004

No 5 “Article IV of the NPT: Background, Problems, Some Prospects” by Lawrence Scheinman, June 2004

No 6 “Nuclear-Weapon-Free Zones: Still a Useful Disarmament and Non-Proliferation Tool?” by Scott Parrish and Jean du Preez, June 2004

No 7 “Making the Non-Proliferation Regime Universal” by Sverre Lodgaard, June 2004

No 8 “Practical Measures to Reduce the Risks Presented by Non-Strategic Nuclear Weapons” by William C. Potter and Nikolai Sokov, June 2004

No 9 “The Future of a Treaty Banning Fissile Material for Weapons Purposes: Is It Still Relevant?” by Jean du Preez, June 2004

No 10 “A Global Assessment of Nuclear Proliferation Threats” by Joseph Cirincione, June 2004

No 11 “Assessing Proposals on the International Nuclear Fuel Cycle” by Jon B. Wolfsthal, June 2004

No 12 “The New Proliferation Game” by William C Potter, June 2004

No 13 “Needed: a Comprehensive Framework for Eliminating WMD” by Michael Krepon, September 2004

No 14 “Managing the Biological Weapons Problem: From the Individual to the International” by Jez Littlewood, August 2004

No 15 “Coping with the Possibility of Terrorist Use of WMD” by Jonathan Dean, June 2004

No 16 “Comparison of States vs. Non-State Actors in the Development of a BTW Capability” by Åke Sellström and Anders Norqvist, October 2004

No 17 “Deconflating ‘WMD’” by George Perkovich, October 2004

No 18 “Global Governance of ‘Contentious’” Science: The Case of the World Health Organization’s Oversight of Small Pox Virus Research” by Jonathan B. Tucker and Stacy M. Okutani, October 2004

No 19 “WMD Verification and Compliance: The State of Play” submitted by Foreign Affairs Canada and prepared by Vertic, October 2004

No 20 “WMD Verification and Compliance: Challenges and Responses” submitted by Foreign Affairs Canada, October 2004

No 21 “Meeting Iran’s Nuclear Challenge” by Gary Samore, October 2004

No 22 “Bioterrorism and Threat Assessment” by Gary A. Ackerman and Kevin S. Moran, November 2004

No 23 “Enhancing BWC Implementation: A Modular Approach” by Trevor Findlay and Angela Woodward, December 2004

No 24 “Controlling Missiles”, by Jonathan Dean, December 2004

No 25 “On Not Confusing the Unfamiliar with the Improbable: Low-Technology Means of Delivering Weapons of Mass Destruction” by Dennis M. Gormley, December 2004

No 26 “A Verification and Transparency Concept for Technology Transfers under the BTWC” by Jean Pascal Zanders, February 2005

No 27 “Missing Piece and Gordian Knot: Missile Non-Proliferation” by Mark Smith, February 2005

No 28 “The Central Importance of Legally Binding Measures for the Strengthening of the Biological and Toxin Weapons Convention (BTWC)” by Graham S. Pearson, February 2005

No 29 “Russia in the PSI: The Modalities of Russian Participation in the Proliferation Security Initiative” by Alexandre Kaliadine, August 2005

No 30 “Indicators of State and Non-State Offensive Chemical and Biological Programmes” edited by Ingrid Fängmark and Lena Norlander, August 2005

No 31 “The 2005 NPT Review Conference: Reasons and Consequences of Failure and Options for Repair” by Harald Müller, August 2005

No 32 “National Measures to Implement WMD Treaties and Norms: the Need for International Standards and Technical Assistance” by Andreas Persbo and Angela Woodward, August 2005

No 33 “Russia and the Chemical Disarmament Process” by Sergey Oznobistchev and Alexander Saveliev, August 2005

No 34 “Transparency and Secrecy in Nuclear Weapons” by Annette Schaper, August 2005

No 35 “Multilateral Nuclear Fuel-Cycle Arrangements” by Harald Müller, August, 2005

No 36 “Nuclear Threat Perceptions and Nonproliferation Responses: A Comparative Analysis” by Scott Parrish and William C. Potter, August, 2005

No 37 “WMD Crisis: Law Instead of Lawless Self-Help” by Harald Müller, August, 2005

No 38 “The Relevance of Gender for Eliminating Weapons of Mass Destruction” by Carol Cohn with Felicity Hill and Sara Ruddick, December, 2005

No 39 “The Influence of the International Trade of Nuclear Materials and Technologies on the Nuclear Nonproliferation Regime” by Dr Vladimir V Evseev, December, 2005

No 40 “A Standing United Nations Verification Body: Necessary and Feasible” by Trevor Findlay, December, 2005

No 41 “Learning from Past Success: The NPT and the Future of Nonproliferation” by Jim Walsh, September, 2006

List of published studies and papersAll papers and studies are available as pdf-files at the Commission’s website: www.wmdcommission.org

THE WEAPONS OF MASS DESTRUCTION COMMISSION

www.wmdcommission.org

jim walsh - kirsch foundation · lessons from success: the npt and the future of non-proliferation...

Documents